* [PATCH v3 net-next 01/13] octeontx2-af: Modify default KEX profile to extract TX packet fields
2020-11-11 7:13 [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Naveen Mamindlapalli
@ 2020-11-11 7:13 ` Naveen Mamindlapalli
2020-11-12 19:57 ` Saeed Mahameed
2020-11-12 20:02 ` Alexander Duyck
2020-11-11 7:13 ` [PATCH v3 net-next 02/13] octeontx2-af: Verify MCAM entry channel and PF_FUNC Naveen Mamindlapalli
` (12 subsequent siblings)
13 siblings, 2 replies; 24+ messages in thread
From: Naveen Mamindlapalli @ 2020-11-11 7:13 UTC (permalink / raw)
To: netdev, linux-kernel
Cc: kuba, davem, saeed, sgoutham, lcherian, gakula, jerinj, sbhatta,
hkelam
From: Stanislaw Kardach <skardach@marvell.com>
The current default Key Extraction(KEX) profile can only use RX
packet fields while generating the MCAM search key. The profile
can't be used for matching TX packet fields. This patch modifies
the default KEX profile to add support for extracting TX packet
fields into MCAM search key. Enabled Tx KPU packet parsing by
configuring TX PKIND in tx_parse_cfg.
Also modified the default KEX profile to extract VLAN TCI from
the LB_PTR and exact byte offset of VLAN header. The NPC KPU
parser was modified to point LB_PTR to the starting byte offset
of VLAN header which points to the tpid field.
Signed-off-by: Stanislaw Kardach <skardach@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
---
.../ethernet/marvell/octeontx2/af/npc_profile.h | 71 ++++++++++++++++++++--
.../net/ethernet/marvell/octeontx2/af/rvu_nix.c | 6 ++
2 files changed, 72 insertions(+), 5 deletions(-)
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/npc_profile.h b/drivers/net/ethernet/marvell/octeontx2/af/npc_profile.h
index 199448610e3e..c5b13385c81d 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/npc_profile.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/npc_profile.h
@@ -13386,8 +13386,8 @@ static struct npc_mcam_kex npc_mkex_default = {
.kpu_version = NPC_KPU_PROFILE_VER,
.keyx_cfg = {
/* nibble: LA..LE (ltype only) + Channel */
- [NIX_INTF_RX] = ((u64)NPC_MCAM_KEY_X2 << 32) | 0x49247,
- [NIX_INTF_TX] = ((u64)NPC_MCAM_KEY_X2 << 32) | ((1ULL << 19) - 1),
+ [NIX_INTF_RX] = ((u64)NPC_MCAM_KEY_X2 << 32) | 0x249207,
+ [NIX_INTF_TX] = ((u64)NPC_MCAM_KEY_X2 << 32) | 0x249200,
},
.intf_lid_lt_ld = {
/* Default RX MCAM KEX profile */
@@ -13405,12 +13405,14 @@ static struct npc_mcam_kex npc_mkex_default = {
/* Layer B: Single VLAN (CTAG) */
/* CTAG VLAN[2..3] + Ethertype, 4 bytes, KW0[63:32] */
[NPC_LT_LB_CTAG] = {
- KEX_LD_CFG(0x03, 0x0, 0x1, 0x0, 0x4),
+ KEX_LD_CFG(0x03, 0x2, 0x1, 0x0, 0x4),
},
/* Layer B: Stacked VLAN (STAG|QinQ) */
[NPC_LT_LB_STAG_QINQ] = {
- /* CTAG VLAN[2..3] + Ethertype, 4 bytes, KW0[63:32] */
- KEX_LD_CFG(0x03, 0x4, 0x1, 0x0, 0x4),
+ /* Outer VLAN: 2 bytes, KW0[63:48] */
+ KEX_LD_CFG(0x01, 0x2, 0x1, 0x0, 0x6),
+ /* Ethertype: 2 bytes, KW0[47:32] */
+ KEX_LD_CFG(0x01, 0x8, 0x1, 0x0, 0x4),
},
[NPC_LT_LB_FDSA] = {
/* SWITCH PORT: 1 byte, KW0[63:48] */
@@ -13450,6 +13452,65 @@ static struct npc_mcam_kex npc_mkex_default = {
},
},
},
+
+ /* Default TX MCAM KEX profile */
+ [NIX_INTF_TX] = {
+ [NPC_LID_LA] = {
+ /* Layer A: Ethernet: */
+ [NPC_LT_LA_IH_NIX_ETHER] = {
+ /* PF_FUNC: 2B , KW0 [47:32] */
+ KEX_LD_CFG(0x01, 0x0, 0x1, 0x0, 0x4),
+ /* DMAC: 6 bytes, KW1[63:16] */
+ KEX_LD_CFG(0x05, 0x8, 0x1, 0x0, 0xa),
+ },
+ },
+ [NPC_LID_LB] = {
+ /* Layer B: Single VLAN (CTAG) */
+ [NPC_LT_LB_CTAG] = {
+ /* CTAG VLAN[2..3] KW0[63:48] */
+ KEX_LD_CFG(0x01, 0x2, 0x1, 0x0, 0x6),
+ /* CTAG VLAN[2..3] KW1[15:0] */
+ KEX_LD_CFG(0x01, 0x4, 0x1, 0x0, 0x8),
+ },
+ /* Layer B: Stacked VLAN (STAG|QinQ) */
+ [NPC_LT_LB_STAG_QINQ] = {
+ /* Outer VLAN: 2 bytes, KW0[63:48] */
+ KEX_LD_CFG(0x01, 0x2, 0x1, 0x0, 0x6),
+ /* Outer VLAN: 2 Bytes, KW1[15:0] */
+ KEX_LD_CFG(0x01, 0x8, 0x1, 0x0, 0x8),
+ },
+ },
+ [NPC_LID_LC] = {
+ /* Layer C: IPv4 */
+ [NPC_LT_LC_IP] = {
+ /* SIP+DIP: 8 bytes, KW2[63:0] */
+ KEX_LD_CFG(0x07, 0xc, 0x1, 0x0, 0x10),
+ /* TOS: 1 byte, KW1[63:56] */
+ KEX_LD_CFG(0x0, 0x1, 0x1, 0x0, 0xf),
+ },
+ /* Layer C: IPv6 */
+ [NPC_LT_LC_IP6] = {
+ /* Everything up to SADDR: 8 bytes, KW2[63:0] */
+ KEX_LD_CFG(0x07, 0x0, 0x1, 0x0, 0x10),
+ },
+ },
+ [NPC_LID_LD] = {
+ /* Layer D:UDP */
+ [NPC_LT_LD_UDP] = {
+ /* SPORT: 2 bytes, KW3[15:0] */
+ KEX_LD_CFG(0x1, 0x0, 0x1, 0x0, 0x18),
+ /* DPORT: 2 bytes, KW3[31:16] */
+ KEX_LD_CFG(0x1, 0x2, 0x1, 0x0, 0x1a),
+ },
+ /* Layer D:TCP */
+ [NPC_LT_LD_TCP] = {
+ /* SPORT: 2 bytes, KW3[15:0] */
+ KEX_LD_CFG(0x1, 0x0, 0x1, 0x0, 0x18),
+ /* DPORT: 2 bytes, KW3[31:16] */
+ KEX_LD_CFG(0x1, 0x2, 0x1, 0x0, 0x1a),
+ },
+ },
+ },
},
};
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
index 8bac1dd3a1c2..8c11abdbd9d1 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
@@ -57,6 +57,8 @@ enum nix_makr_fmt_indexes {
NIX_MARK_CFG_MAX,
};
+#define NIX_TX_PKIND 63ULL
+
/* For now considering MC resources needed for broadcast
* pkt replication only. i.e 256 HWVFs + 12 PFs.
*/
@@ -1182,6 +1184,10 @@ int rvu_mbox_handler_nix_lf_alloc(struct rvu *rvu,
/* Config Rx pkt length, csum checks and apad enable / disable */
rvu_write64(rvu, blkaddr, NIX_AF_LFX_RX_CFG(nixlf), req->rx_cfg);
+ /* Configure pkind for TX parse config, 63 from npc_profile */
+ cfg = NIX_TX_PKIND;
+ rvu_write64(rvu, blkaddr, NIX_AF_LFX_TX_PARSE_CFG(nixlf), cfg);
+
intf = is_afvf(pcifunc) ? NIX_INTF_TYPE_LBK : NIX_INTF_TYPE_CGX;
err = nix_interface_init(rvu, pcifunc, intf, nixlf);
if (err)
--
2.16.5
^ permalink raw reply related [flat|nested] 24+ messages in thread* Re: [PATCH v3 net-next 01/13] octeontx2-af: Modify default KEX profile to extract TX packet fields
2020-11-11 7:13 ` [PATCH v3 net-next 01/13] octeontx2-af: Modify default KEX profile to extract TX packet fields Naveen Mamindlapalli
@ 2020-11-12 19:57 ` Saeed Mahameed
2020-11-14 18:59 ` [EXT] " Naveen Mamindlapalli
2020-11-12 20:02 ` Alexander Duyck
1 sibling, 1 reply; 24+ messages in thread
From: Saeed Mahameed @ 2020-11-12 19:57 UTC (permalink / raw)
To: Naveen Mamindlapalli, netdev, linux-kernel
Cc: kuba, davem, sgoutham, lcherian, gakula, jerinj, sbhatta, hkelam
On Wed, 2020-11-11 at 12:43 +0530, Naveen Mamindlapalli wrote:
> From: Stanislaw Kardach <skardach@marvell.com>
>
> The current default Key Extraction(KEX) profile can only use RX
> packet fields while generating the MCAM search key. The profile
> can't be used for matching TX packet fields. This patch modifies
> the default KEX profile to add support for extracting TX packet
> fields into MCAM search key. Enabled Tx KPU packet parsing by
> configuring TX PKIND in tx_parse_cfg.
>
> Also modified the default KEX profile to extract VLAN TCI from
> the LB_PTR and exact byte offset of VLAN header. The NPC KPU
> parser was modified to point LB_PTR to the starting byte offset
> of VLAN header which points to the tpid field.
>
> Signed-off-by: Stanislaw Kardach <skardach@marvell.com>
> Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
> Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
^ permalink raw reply [flat|nested] 24+ messages in thread
* RE: [EXT] Re: [PATCH v3 net-next 01/13] octeontx2-af: Modify default KEX profile to extract TX packet fields
2020-11-12 19:57 ` Saeed Mahameed
@ 2020-11-14 18:59 ` Naveen Mamindlapalli
0 siblings, 0 replies; 24+ messages in thread
From: Naveen Mamindlapalli @ 2020-11-14 18:59 UTC (permalink / raw)
To: Saeed Mahameed, netdev@vger.kernel.org,
linux-kernel@vger.kernel.org
Cc: kuba@kernel.org, davem@davemloft.net, Sunil Kovvuri Goutham,
Linu Cherian, Geethasowjanya Akula, Jerin Jacob Kollanukkaran,
Subbaraya Sundeep Bhatta, Hariprasad Kelam
Hi Saeed,
Thanks for the review.
> -----Original Message-----
> From: Saeed Mahameed <saeed@kernel.org>
> Sent: Friday, November 13, 2020 1:27 AM
> To: Naveen Mamindlapalli <naveenm@marvell.com>; netdev@vger.kernel.org;
> linux-kernel@vger.kernel.org
> Cc: kuba@kernel.org; davem@davemloft.net; Sunil Kovvuri Goutham
> <sgoutham@marvell.com>; Linu Cherian <lcherian@marvell.com>;
> Geethasowjanya Akula <gakula@marvell.com>; Jerin Jacob Kollanukkaran
> <jerinj@marvell.com>; Subbaraya Sundeep Bhatta <sbhatta@marvell.com>;
> Hariprasad Kelam <hkelam@marvell.com>
> Subject: [EXT] Re: [PATCH v3 net-next 01/13] octeontx2-af: Modify default KEX
> profile to extract TX packet fields
>
> External Email
>
> ----------------------------------------------------------------------
> On Wed, 2020-11-11 at 12:43 +0530, Naveen Mamindlapalli wrote:
> > From: Stanislaw Kardach <skardach@marvell.com>
> >
> > The current default Key Extraction(KEX) profile can only use RX packet
> > fields while generating the MCAM search key. The profile can't be used
> > for matching TX packet fields. This patch modifies the default KEX
> > profile to add support for extracting TX packet fields into MCAM
> > search key. Enabled Tx KPU packet parsing by configuring TX PKIND in
> > tx_parse_cfg.
> >
> > Also modified the default KEX profile to extract VLAN TCI from the
> > LB_PTR and exact byte offset of VLAN header. The NPC KPU parser was
> > modified to point LB_PTR to the starting byte offset of VLAN header
> > which points to the tpid field.
> >
> > Signed-off-by: Stanislaw Kardach <skardach@marvell.com>
> > Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
> > Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
> >
>
> Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
>
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH v3 net-next 01/13] octeontx2-af: Modify default KEX profile to extract TX packet fields
2020-11-11 7:13 ` [PATCH v3 net-next 01/13] octeontx2-af: Modify default KEX profile to extract TX packet fields Naveen Mamindlapalli
2020-11-12 19:57 ` Saeed Mahameed
@ 2020-11-12 20:02 ` Alexander Duyck
1 sibling, 0 replies; 24+ messages in thread
From: Alexander Duyck @ 2020-11-12 20:02 UTC (permalink / raw)
To: Naveen Mamindlapalli
Cc: Netdev, LKML, Jakub Kicinski, David Miller, saeed, sgoutham,
lcherian, gakula, jerinj, sbhatta, hkelam
On Tue, Nov 10, 2020 at 11:22 PM Naveen Mamindlapalli
<naveenm@marvell.com> wrote:
>
> From: Stanislaw Kardach <skardach@marvell.com>
>
> The current default Key Extraction(KEX) profile can only use RX
> packet fields while generating the MCAM search key. The profile
> can't be used for matching TX packet fields. This patch modifies
> the default KEX profile to add support for extracting TX packet
> fields into MCAM search key. Enabled Tx KPU packet parsing by
> configuring TX PKIND in tx_parse_cfg.
>
> Also modified the default KEX profile to extract VLAN TCI from
> the LB_PTR and exact byte offset of VLAN header. The NPC KPU
> parser was modified to point LB_PTR to the starting byte offset
> of VLAN header which points to the tpid field.
>
> Signed-off-by: Stanislaw Kardach <skardach@marvell.com>
> Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
> Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
A bit more documentation would be useful. However other than that the
code itself appears to make sense.
Reviewed-by: Alexander Duyck <alexanderduyck@fb.com>
> ---
> .../ethernet/marvell/octeontx2/af/npc_profile.h | 71 ++++++++++++++++++++--
> .../net/ethernet/marvell/octeontx2/af/rvu_nix.c | 6 ++
> 2 files changed, 72 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/net/ethernet/marvell/octeontx2/af/npc_profile.h b/drivers/net/ethernet/marvell/octeontx2/af/npc_profile.h
> index 199448610e3e..c5b13385c81d 100644
> --- a/drivers/net/ethernet/marvell/octeontx2/af/npc_profile.h
> +++ b/drivers/net/ethernet/marvell/octeontx2/af/npc_profile.h
> @@ -13386,8 +13386,8 @@ static struct npc_mcam_kex npc_mkex_default = {
> .kpu_version = NPC_KPU_PROFILE_VER,
> .keyx_cfg = {
> /* nibble: LA..LE (ltype only) + Channel */
> - [NIX_INTF_RX] = ((u64)NPC_MCAM_KEY_X2 << 32) | 0x49247,
> - [NIX_INTF_TX] = ((u64)NPC_MCAM_KEY_X2 << 32) | ((1ULL << 19) - 1),
> + [NIX_INTF_RX] = ((u64)NPC_MCAM_KEY_X2 << 32) | 0x249207,
> + [NIX_INTF_TX] = ((u64)NPC_MCAM_KEY_X2 << 32) | 0x249200,
> },
> .intf_lid_lt_ld = {
> /* Default RX MCAM KEX profile */
//
Any sort of explanation for what some of these magic numbers means
might be useful. I'm left wondering if the lower 32b is a bitfield or
a fixed value. I am guessing bit field based on the fact that it was
originally being set using ((1ULL << X) - 1) however if there were
macros defined for each bit explaining what each bit was that would be
useful.
> @@ -13405,12 +13405,14 @@ static struct npc_mcam_kex npc_mkex_default = {
> /* Layer B: Single VLAN (CTAG) */
> /* CTAG VLAN[2..3] + Ethertype, 4 bytes, KW0[63:32] */
> [NPC_LT_LB_CTAG] = {
> - KEX_LD_CFG(0x03, 0x0, 0x1, 0x0, 0x4),
> + KEX_LD_CFG(0x03, 0x2, 0x1, 0x0, 0x4),
> },
Similarly here some explanation for KEX_LD_CFG would be useful. From
what I can tell it seems like this may be some sort of fix as you are
adjusting the "hdr_ofs" field from 0 to 2.
> /* Layer B: Stacked VLAN (STAG|QinQ) */
> [NPC_LT_LB_STAG_QINQ] = {
> - /* CTAG VLAN[2..3] + Ethertype, 4 bytes, KW0[63:32] */
> - KEX_LD_CFG(0x03, 0x4, 0x1, 0x0, 0x4),
> + /* Outer VLAN: 2 bytes, KW0[63:48] */
> + KEX_LD_CFG(0x01, 0x2, 0x1, 0x0, 0x6),
> + /* Ethertype: 2 bytes, KW0[47:32] */
> + KEX_LD_CFG(0x01, 0x8, 0x1, 0x0, 0x4),
Just to confirm, are you matching up the outer VLAN with the inner
Ethertype here? It seems like an odd combination. I assume you need
the inner ethertype in order to identify the L3 traffic?
> },
> [NPC_LT_LB_FDSA] = {
> /* SWITCH PORT: 1 byte, KW0[63:48] */
> @@ -13450,6 +13452,65 @@ static struct npc_mcam_kex npc_mkex_default = {
> },
> },
> },
> +
> + /* Default TX MCAM KEX profile */
> + [NIX_INTF_TX] = {
> + [NPC_LID_LA] = {
> + /* Layer A: Ethernet: */
> + [NPC_LT_LA_IH_NIX_ETHER] = {
> + /* PF_FUNC: 2B , KW0 [47:32] */
> + KEX_LD_CFG(0x01, 0x0, 0x1, 0x0, 0x4),
I'm assuming you have an 8B internal header that is being parsed? A
comment explaining that this is parsing a preamble that is at the
start of things might be useful.
> + /* DMAC: 6 bytes, KW1[63:16] */
> + KEX_LD_CFG(0x05, 0x8, 0x1, 0x0, 0xa),
> + },
> + },
> + [NPC_LID_LB] = {
> + /* Layer B: Single VLAN (CTAG) */
> + [NPC_LT_LB_CTAG] = {
> + /* CTAG VLAN[2..3] KW0[63:48] */
> + KEX_LD_CFG(0x01, 0x2, 0x1, 0x0, 0x6),
> + /* CTAG VLAN[2..3] KW1[15:0] */
> + KEX_LD_CFG(0x01, 0x4, 0x1, 0x0, 0x8),
> + },
> + /* Layer B: Stacked VLAN (STAG|QinQ) */
> + [NPC_LT_LB_STAG_QINQ] = {
> + /* Outer VLAN: 2 bytes, KW0[63:48] */
> + KEX_LD_CFG(0x01, 0x2, 0x1, 0x0, 0x6),
> + /* Outer VLAN: 2 Bytes, KW1[15:0] */
> + KEX_LD_CFG(0x01, 0x8, 0x1, 0x0, 0x8),
> + },
> + },
> + [NPC_LID_LC] = {
> + /* Layer C: IPv4 */
> + [NPC_LT_LC_IP] = {
> + /* SIP+DIP: 8 bytes, KW2[63:0] */
> + KEX_LD_CFG(0x07, 0xc, 0x1, 0x0, 0x10),
> + /* TOS: 1 byte, KW1[63:56] */
> + KEX_LD_CFG(0x0, 0x1, 0x1, 0x0, 0xf),
> + },
> + /* Layer C: IPv6 */
> + [NPC_LT_LC_IP6] = {
> + /* Everything up to SADDR: 8 bytes, KW2[63:0] */
> + KEX_LD_CFG(0x07, 0x0, 0x1, 0x0, 0x10),
> + },
> + },
> + [NPC_LID_LD] = {
> + /* Layer D:UDP */
> + [NPC_LT_LD_UDP] = {
> + /* SPORT: 2 bytes, KW3[15:0] */
> + KEX_LD_CFG(0x1, 0x0, 0x1, 0x0, 0x18),
> + /* DPORT: 2 bytes, KW3[31:16] */
> + KEX_LD_CFG(0x1, 0x2, 0x1, 0x0, 0x1a),
I am curious why this is being done as two pieces instead of just one.
From what I can tell you could just have a single rule that copies the
4 bytes for both ports in one shot and they would end up in the same
place wouldn't they?
> + },
> + /* Layer D:TCP */
> + [NPC_LT_LD_TCP] = {
> + /* SPORT: 2 bytes, KW3[15:0] */
> + KEX_LD_CFG(0x1, 0x0, 0x1, 0x0, 0x18),
> + /* DPORT: 2 bytes, KW3[31:16] */
> + KEX_LD_CFG(0x1, 0x2, 0x1, 0x0, 0x1a),
> + },
> + },
> + },
> },
> };
>
> diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
> index 8bac1dd3a1c2..8c11abdbd9d1 100644
> --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
> +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
> @@ -57,6 +57,8 @@ enum nix_makr_fmt_indexes {
> NIX_MARK_CFG_MAX,
> };
>
> +#define NIX_TX_PKIND 63ULL
> +
A comment explaining the magic number would be useful. From what I can
tell this looks like a "just turn everything on" sort of config where
you are enabling bit flags 0 - 5.
> /* For now considering MC resources needed for broadcast
> * pkt replication only. i.e 256 HWVFs + 12 PFs.
> */
> @@ -1182,6 +1184,10 @@ int rvu_mbox_handler_nix_lf_alloc(struct rvu *rvu,
> /* Config Rx pkt length, csum checks and apad enable / disable */
> rvu_write64(rvu, blkaddr, NIX_AF_LFX_RX_CFG(nixlf), req->rx_cfg);
>
> + /* Configure pkind for TX parse config, 63 from npc_profile */
> + cfg = NIX_TX_PKIND;
> + rvu_write64(rvu, blkaddr, NIX_AF_LFX_TX_PARSE_CFG(nixlf), cfg);
> +
> intf = is_afvf(pcifunc) ? NIX_INTF_TYPE_LBK : NIX_INTF_TYPE_CGX;
> err = nix_interface_init(rvu, pcifunc, intf, nixlf);
> if (err)
> --
> 2.16.5
>
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH v3 net-next 02/13] octeontx2-af: Verify MCAM entry channel and PF_FUNC
2020-11-11 7:13 [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Naveen Mamindlapalli
2020-11-11 7:13 ` [PATCH v3 net-next 01/13] octeontx2-af: Modify default KEX profile to extract TX packet fields Naveen Mamindlapalli
@ 2020-11-11 7:13 ` Naveen Mamindlapalli
2020-11-12 20:03 ` Alexander Duyck
2020-11-11 7:13 ` [PATCH v3 net-next 03/13] octeontx2-af: Generate key field bit mask from KEX profile Naveen Mamindlapalli
` (11 subsequent siblings)
13 siblings, 1 reply; 24+ messages in thread
From: Naveen Mamindlapalli @ 2020-11-11 7:13 UTC (permalink / raw)
To: netdev, linux-kernel
Cc: kuba, davem, saeed, sgoutham, lcherian, gakula, jerinj, sbhatta,
hkelam
From: Subbaraya Sundeep <sbhatta@marvell.com>
This patch adds support to verify the channel number sent by
mailbox requester before writing MCAM entry for Ingress packets.
Similarly for Egress packets, verifying the PF_FUNC sent by the
mailbox user.
Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: Kiran Kumar K <kirankumark@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
---
drivers/net/ethernet/marvell/octeontx2/af/rvu.c | 4 +-
drivers/net/ethernet/marvell/octeontx2/af/rvu.h | 2 +
.../net/ethernet/marvell/octeontx2/af/rvu_npc.c | 78 ++++++++++++++++++++++
3 files changed, 82 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu.c
index a28a518c0eae..e8b5aaf73201 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.c
@@ -2642,7 +2642,7 @@ static void rvu_enable_afvf_intr(struct rvu *rvu)
#define PCI_DEVID_OCTEONTX2_LBK 0xA061
-static int lbk_get_num_chans(void)
+int rvu_get_num_lbk_chans(void)
{
struct pci_dev *pdev;
void __iomem *base;
@@ -2677,7 +2677,7 @@ static int rvu_enable_sriov(struct rvu *rvu)
return 0;
}
- chans = lbk_get_num_chans();
+ chans = rvu_get_num_lbk_chans();
if (chans < 0)
return chans;
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
index 5ac9bb12415f..1724dbd18847 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
@@ -445,6 +445,7 @@ int rvu_get_lf(struct rvu *rvu, struct rvu_block *block, u16 pcifunc, u16 slot);
int rvu_lf_reset(struct rvu *rvu, struct rvu_block *block, int lf);
int rvu_get_blkaddr(struct rvu *rvu, int blktype, u16 pcifunc);
int rvu_poll_reg(struct rvu *rvu, u64 block, u64 offset, u64 mask, bool zero);
+int rvu_get_num_lbk_chans(void);
/* RVU HW reg validation */
enum regmap_block {
@@ -535,6 +536,7 @@ bool is_npc_intf_tx(u8 intf);
bool is_npc_intf_rx(u8 intf);
bool is_npc_interface_valid(struct rvu *rvu, u8 intf);
int rvu_npc_get_tx_nibble_cfg(struct rvu *rvu, u64 nibble_ena);
+int npc_mcam_verify_channel(struct rvu *rvu, u16 pcifunc, u8 intf, u16 channel);
#ifdef CONFIG_DEBUG_FS
void rvu_dbg_init(struct rvu *rvu);
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
index 989533a3d2ce..3666159bb6b6 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
@@ -28,6 +28,8 @@
#define NPC_PARSE_RESULT_DMAC_OFFSET 8
#define NPC_HW_TSTAMP_OFFSET 8
+#define NPC_KEX_CHAN_MASK 0xFFFULL
+#define NPC_KEX_PF_FUNC_MASK 0xFFFFULL
static const char def_pfl_name[] = "default";
@@ -63,6 +65,54 @@ int rvu_npc_get_tx_nibble_cfg(struct rvu *rvu, u64 nibble_ena)
return 0;
}
+static int npc_mcam_verify_pf_func(struct rvu *rvu,
+ struct mcam_entry *entry_data, u8 intf,
+ u16 pcifunc)
+{
+ u16 pf_func, pf_func_mask;
+
+ if (is_npc_intf_rx(intf))
+ return 0;
+
+ pf_func_mask = (entry_data->kw_mask[0] >> 32) &
+ NPC_KEX_PF_FUNC_MASK;
+ pf_func = (entry_data->kw[0] >> 32) & NPC_KEX_PF_FUNC_MASK;
+
+ pf_func = be16_to_cpu((__force __be16)pf_func);
+ if (pf_func_mask != NPC_KEX_PF_FUNC_MASK ||
+ ((pf_func & ~RVU_PFVF_FUNC_MASK) !=
+ (pcifunc & ~RVU_PFVF_FUNC_MASK)))
+ return -EINVAL;
+
+ return 0;
+}
+
+int npc_mcam_verify_channel(struct rvu *rvu, u16 pcifunc, u8 intf, u16 channel)
+{
+ int pf = rvu_get_pf(pcifunc);
+ u8 cgx_id, lmac_id;
+ int base = 0, end;
+
+ if (is_npc_intf_tx(intf))
+ return 0;
+
+ if (is_afvf(pcifunc)) {
+ end = rvu_get_num_lbk_chans();
+ if (end < 0)
+ return -EINVAL;
+ } else {
+ rvu_get_cgx_lmac_id(rvu->pf2cgxlmac_map[pf], &cgx_id, &lmac_id);
+ base = NIX_CHAN_CGX_LMAC_CHX(cgx_id, lmac_id, 0x0);
+ /* CGX mapped functions has maximum of 16 channels */
+ end = NIX_CHAN_CGX_LMAC_CHX(cgx_id, lmac_id, 0xF);
+ }
+
+ if (channel < base || channel > end)
+ return -EINVAL;
+
+ return 0;
+}
+
void rvu_npc_set_pkind(struct rvu *rvu, int pkind, struct rvu_pfvf *pfvf)
{
int blkaddr;
@@ -1935,6 +1985,7 @@ int rvu_mbox_handler_npc_mcam_write_entry(struct rvu *rvu,
struct rvu_pfvf *pfvf = rvu_get_pfvf(rvu, req->hdr.pcifunc);
struct npc_mcam *mcam = &rvu->hw->mcam;
u16 pcifunc = req->hdr.pcifunc;
+ u16 channel, chan_mask;
int blkaddr, rc;
u8 nix_intf;
@@ -1942,6 +1993,10 @@ int rvu_mbox_handler_npc_mcam_write_entry(struct rvu *rvu,
if (blkaddr < 0)
return NPC_MCAM_INVALID_REQ;
+ chan_mask = req->entry_data.kw_mask[0] & NPC_KEX_CHAN_MASK;
+ channel = req->entry_data.kw[0] & NPC_KEX_CHAN_MASK;
+ channel &= chan_mask;
+
mutex_lock(&mcam->lock);
rc = npc_mcam_verify_entry(mcam, pcifunc, req->entry);
if (rc)
@@ -1963,6 +2018,17 @@ int rvu_mbox_handler_npc_mcam_write_entry(struct rvu *rvu,
else
nix_intf = pfvf->nix_rx_intf;
+ if (npc_mcam_verify_channel(rvu, pcifunc, req->intf, channel)) {
+ rc = NPC_MCAM_INVALID_REQ;
+ goto exit;
+ }
+
+ if (npc_mcam_verify_pf_func(rvu, &req->entry_data, req->intf,
+ pcifunc)) {
+ rc = NPC_MCAM_INVALID_REQ;
+ goto exit;
+ }
+
npc_config_mcam_entry(rvu, mcam, blkaddr, req->entry, nix_intf,
&req->entry_data, req->enable_entry);
@@ -2299,6 +2365,7 @@ int rvu_mbox_handler_npc_mcam_alloc_and_write_entry(struct rvu *rvu,
struct npc_mcam *mcam = &rvu->hw->mcam;
u16 entry = NPC_MCAM_ENTRY_INVALID;
u16 cntr = NPC_MCAM_ENTRY_INVALID;
+ u16 channel, chan_mask;
int blkaddr, rc;
u8 nix_intf;
@@ -2309,6 +2376,17 @@ int rvu_mbox_handler_npc_mcam_alloc_and_write_entry(struct rvu *rvu,
if (!is_npc_interface_valid(rvu, req->intf))
return NPC_MCAM_INVALID_REQ;
+ chan_mask = req->entry_data.kw_mask[0] & NPC_KEX_CHAN_MASK;
+ channel = req->entry_data.kw[0] & NPC_KEX_CHAN_MASK;
+ channel &= chan_mask;
+
+ if (npc_mcam_verify_channel(rvu, req->hdr.pcifunc, req->intf, channel))
+ return NPC_MCAM_INVALID_REQ;
+
+ if (npc_mcam_verify_pf_func(rvu, &req->entry_data, req->intf,
+ req->hdr.pcifunc))
+ return NPC_MCAM_INVALID_REQ;
+
/* Try to allocate a MCAM entry */
entry_req.hdr.pcifunc = req->hdr.pcifunc;
entry_req.contig = true;
--
2.16.5
^ permalink raw reply related [flat|nested] 24+ messages in thread* Re: [PATCH v3 net-next 02/13] octeontx2-af: Verify MCAM entry channel and PF_FUNC
2020-11-11 7:13 ` [PATCH v3 net-next 02/13] octeontx2-af: Verify MCAM entry channel and PF_FUNC Naveen Mamindlapalli
@ 2020-11-12 20:03 ` Alexander Duyck
0 siblings, 0 replies; 24+ messages in thread
From: Alexander Duyck @ 2020-11-12 20:03 UTC (permalink / raw)
To: Naveen Mamindlapalli
Cc: Netdev, LKML, Jakub Kicinski, David Miller, saeed, sgoutham,
lcherian, gakula, jerinj, sbhatta, hkelam
On Tue, Nov 10, 2020 at 11:18 PM Naveen Mamindlapalli
<naveenm@marvell.com> wrote:
>
> From: Subbaraya Sundeep <sbhatta@marvell.com>
>
> This patch adds support to verify the channel number sent by
> mailbox requester before writing MCAM entry for Ingress packets.
> Similarly for Egress packets, verifying the PF_FUNC sent by the
> mailbox user.
>
> Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
> Signed-off-by: Kiran Kumar K <kirankumark@marvell.com>
> Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
> Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
One minor nit below. Otherwise looks good to me.
Reviewed-by: Alexander Duyck <alexanderduyck@fb.com>
> ---
> drivers/net/ethernet/marvell/octeontx2/af/rvu.c | 4 +-
> drivers/net/ethernet/marvell/octeontx2/af/rvu.h | 2 +
> .../net/ethernet/marvell/octeontx2/af/rvu_npc.c | 78 ++++++++++++++++++++++
> 3 files changed, 82 insertions(+), 2 deletions(-)
>
> diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu.c
> index a28a518c0eae..e8b5aaf73201 100644
> --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.c
> +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.c
> @@ -2642,7 +2642,7 @@ static void rvu_enable_afvf_intr(struct rvu *rvu)
>
> #define PCI_DEVID_OCTEONTX2_LBK 0xA061
>
> -static int lbk_get_num_chans(void)
> +int rvu_get_num_lbk_chans(void)
> {
> struct pci_dev *pdev;
> void __iomem *base;
> @@ -2677,7 +2677,7 @@ static int rvu_enable_sriov(struct rvu *rvu)
> return 0;
> }
>
> - chans = lbk_get_num_chans();
> + chans = rvu_get_num_lbk_chans();
> if (chans < 0)
> return chans;
>
> diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
> index 5ac9bb12415f..1724dbd18847 100644
> --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
> +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
> @@ -445,6 +445,7 @@ int rvu_get_lf(struct rvu *rvu, struct rvu_block *block, u16 pcifunc, u16 slot);
> int rvu_lf_reset(struct rvu *rvu, struct rvu_block *block, int lf);
> int rvu_get_blkaddr(struct rvu *rvu, int blktype, u16 pcifunc);
> int rvu_poll_reg(struct rvu *rvu, u64 block, u64 offset, u64 mask, bool zero);
> +int rvu_get_num_lbk_chans(void);
>
> /* RVU HW reg validation */
> enum regmap_block {
> @@ -535,6 +536,7 @@ bool is_npc_intf_tx(u8 intf);
> bool is_npc_intf_rx(u8 intf);
> bool is_npc_interface_valid(struct rvu *rvu, u8 intf);
> int rvu_npc_get_tx_nibble_cfg(struct rvu *rvu, u64 nibble_ena);
> +int npc_mcam_verify_channel(struct rvu *rvu, u16 pcifunc, u8 intf, u16 channel);
>
> #ifdef CONFIG_DEBUG_FS
> void rvu_dbg_init(struct rvu *rvu);
> diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
> index 989533a3d2ce..3666159bb6b6 100644
> --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
> +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
> @@ -28,6 +28,8 @@
>
> #define NPC_PARSE_RESULT_DMAC_OFFSET 8
> #define NPC_HW_TSTAMP_OFFSET 8
> +#define NPC_KEX_CHAN_MASK 0xFFFULL
> +#define NPC_KEX_PF_FUNC_MASK 0xFFFFULL
>
> static const char def_pfl_name[] = "default";
>
> @@ -63,6 +65,54 @@ int rvu_npc_get_tx_nibble_cfg(struct rvu *rvu, u64 nibble_ena)
> return 0;
> }
>
> +static int npc_mcam_verify_pf_func(struct rvu *rvu,
> + struct mcam_entry *entry_data, u8 intf,
> + u16 pcifunc)
> +{
> + u16 pf_func, pf_func_mask;
> +
> + if (is_npc_intf_rx(intf))
> + return 0;
> +
> + pf_func_mask = (entry_data->kw_mask[0] >> 32) &
> + NPC_KEX_PF_FUNC_MASK;
> + pf_func = (entry_data->kw[0] >> 32) & NPC_KEX_PF_FUNC_MASK;
> +
> + pf_func = be16_to_cpu((__force __be16)pf_func);
> + if (pf_func_mask != NPC_KEX_PF_FUNC_MASK ||
> + ((pf_func & ~RVU_PFVF_FUNC_MASK) !=
> + (pcifunc & ~RVU_PFVF_FUNC_MASK)))
> + return -EINVAL;
> +
> + return 0;
> +}
> +
> +int npc_mcam_verify_channel(struct rvu *rvu, u16 pcifunc, u8 intf, u16 channel)
> +{
> + int pf = rvu_get_pf(pcifunc);
> + u8 cgx_id, lmac_id;
> + int base = 0, end;
> +
> + if (is_npc_intf_tx(intf))
> + return 0;
> +
> + if (is_afvf(pcifunc)) {
> + end = rvu_get_num_lbk_chans();
> + if (end < 0)
> + return -EINVAL;
> + } else {
> + rvu_get_cgx_lmac_id(rvu->pf2cgxlmac_map[pf], &cgx_id, &lmac_id);
> + base = NIX_CHAN_CGX_LMAC_CHX(cgx_id, lmac_id, 0x0);
> + /* CGX mapped functions has maximum of 16 channels */
> + end = NIX_CHAN_CGX_LMAC_CHX(cgx_id, lmac_id, 0xF);
> + }
> +
> + if (channel < base || channel > end)
> + return -EINVAL;
> +
> + return 0;
> +}
> +
> void rvu_npc_set_pkind(struct rvu *rvu, int pkind, struct rvu_pfvf *pfvf)
> {
> int blkaddr;
> @@ -1935,6 +1985,7 @@ int rvu_mbox_handler_npc_mcam_write_entry(struct rvu *rvu,
> struct rvu_pfvf *pfvf = rvu_get_pfvf(rvu, req->hdr.pcifunc);
> struct npc_mcam *mcam = &rvu->hw->mcam;
> u16 pcifunc = req->hdr.pcifunc;
> + u16 channel, chan_mask;
> int blkaddr, rc;
> u8 nix_intf;
>
> @@ -1942,6 +1993,10 @@ int rvu_mbox_handler_npc_mcam_write_entry(struct rvu *rvu,
> if (blkaddr < 0)
> return NPC_MCAM_INVALID_REQ;
>
> + chan_mask = req->entry_data.kw_mask[0] & NPC_KEX_CHAN_MASK;
> + channel = req->entry_data.kw[0] & NPC_KEX_CHAN_MASK;
> + channel &= chan_mask;
> +
> mutex_lock(&mcam->lock);
> rc = npc_mcam_verify_entry(mcam, pcifunc, req->entry);
> if (rc)
> @@ -1963,6 +2018,17 @@ int rvu_mbox_handler_npc_mcam_write_entry(struct rvu *rvu,
> else
> nix_intf = pfvf->nix_rx_intf;
>
> + if (npc_mcam_verify_channel(rvu, pcifunc, req->intf, channel)) {
> + rc = NPC_MCAM_INVALID_REQ;
> + goto exit;
> + }
> +
> + if (npc_mcam_verify_pf_func(rvu, &req->entry_data, req->intf,
> + pcifunc)) {
> + rc = NPC_MCAM_INVALID_REQ;
> + goto exit;
> + }
> +
> npc_config_mcam_entry(rvu, mcam, blkaddr, req->entry, nix_intf,
> &req->entry_data, req->enable_entry);
>
> @@ -2299,6 +2365,7 @@ int rvu_mbox_handler_npc_mcam_alloc_and_write_entry(struct rvu *rvu,
> struct npc_mcam *mcam = &rvu->hw->mcam;
> u16 entry = NPC_MCAM_ENTRY_INVALID;
> u16 cntr = NPC_MCAM_ENTRY_INVALID;
> + u16 channel, chan_mask;
> int blkaddr, rc;
> u8 nix_intf;
>
> @@ -2309,6 +2376,17 @@ int rvu_mbox_handler_npc_mcam_alloc_and_write_entry(struct rvu *rvu,
> if (!is_npc_interface_valid(rvu, req->intf))
> return NPC_MCAM_INVALID_REQ;
>
> + chan_mask = req->entry_data.kw_mask[0] & NPC_KEX_CHAN_MASK;
> + channel = req->entry_data.kw[0] & NPC_KEX_CHAN_MASK;
> + channel &= chan_mask;
> +
> + if (npc_mcam_verify_channel(rvu, req->hdr.pcifunc, req->intf, channel))
> + return NPC_MCAM_INVALID_REQ;
> +
Why not just move the code for pulling the channel into the
npc_mcam_verify_channel function like you did with the pf_func? Then
you can avoid declaring variables here that won't be used anywhere
else in the function.
> + if (npc_mcam_verify_pf_func(rvu, &req->entry_data, req->intf,
> + req->hdr.pcifunc))
> + return NPC_MCAM_INVALID_REQ;
> +
> /* Try to allocate a MCAM entry */
> entry_req.hdr.pcifunc = req->hdr.pcifunc;
> entry_req.contig = true;
> --
> 2.16.5
>
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH v3 net-next 03/13] octeontx2-af: Generate key field bit mask from KEX profile
2020-11-11 7:13 [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Naveen Mamindlapalli
2020-11-11 7:13 ` [PATCH v3 net-next 01/13] octeontx2-af: Modify default KEX profile to extract TX packet fields Naveen Mamindlapalli
2020-11-11 7:13 ` [PATCH v3 net-next 02/13] octeontx2-af: Verify MCAM entry channel and PF_FUNC Naveen Mamindlapalli
@ 2020-11-11 7:13 ` Naveen Mamindlapalli
2020-11-12 21:24 ` Alexander Duyck
2020-11-11 7:13 ` [PATCH v3 net-next 04/13] octeontx2-af: Add mbox messages to install and delete MCAM rules Naveen Mamindlapalli
` (10 subsequent siblings)
13 siblings, 1 reply; 24+ messages in thread
From: Naveen Mamindlapalli @ 2020-11-11 7:13 UTC (permalink / raw)
To: netdev, linux-kernel
Cc: kuba, davem, saeed, sgoutham, lcherian, gakula, jerinj, sbhatta,
hkelam
From: Subbaraya Sundeep <sbhatta@marvell.com>
Key Extraction(KEX) profile decides how the packet metadata such as
layer information and selected packet data bytes at each layer are
placed in MCAM search key. This patch reads the configured KEX profile
parameters to find out the bit position and bit mask for each field.
The information is used when programming the MCAM match data by SW
to match a packet flow and take appropriate action on the flow. This
patch also verifies the mandatory fields such as channel and DMAC
are not overwritten by the KEX configuration of other fields.
Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
---
drivers/net/ethernet/marvell/octeontx2/af/Makefile | 2 +-
drivers/net/ethernet/marvell/octeontx2/af/npc.h | 48 ++
drivers/net/ethernet/marvell/octeontx2/af/rvu.h | 38 ++
.../net/ethernet/marvell/octeontx2/af/rvu_npc.c | 11 +-
.../net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c | 562 +++++++++++++++++++++
5 files changed, 658 insertions(+), 3 deletions(-)
create mode 100644 drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/Makefile b/drivers/net/ethernet/marvell/octeontx2/af/Makefile
index 2f7a861d0c7b..ffc681b67f1c 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/Makefile
+++ b/drivers/net/ethernet/marvell/octeontx2/af/Makefile
@@ -9,4 +9,4 @@ obj-$(CONFIG_OCTEONTX2_AF) += octeontx2_af.o
octeontx2_mbox-y := mbox.o rvu_trace.o
octeontx2_af-y := cgx.o rvu.o rvu_cgx.o rvu_npa.o rvu_nix.o \
- rvu_reg.o rvu_npc.o rvu_debugfs.o ptp.o
+ rvu_reg.o rvu_npc.o rvu_debugfs.o ptp.o rvu_npc_fs.o
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/npc.h b/drivers/net/ethernet/marvell/octeontx2/af/npc.h
index 91a9d00e4fb5..0fe47216f771 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/npc.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/npc.h
@@ -140,6 +140,54 @@ enum npc_kpu_lh_ltype {
NPC_LT_LH_CUSTOM1 = 0xF,
};
+/* list of known and supported fields in packet header and
+ * fields present in key structure.
+ */
+enum key_fields {
+ NPC_DMAC,
+ NPC_SMAC,
+ NPC_ETYPE,
+ NPC_OUTER_VID,
+ NPC_TOS,
+ NPC_SIP_IPV4,
+ NPC_DIP_IPV4,
+ NPC_SIP_IPV6,
+ NPC_DIP_IPV6,
+ NPC_SPORT_TCP,
+ NPC_DPORT_TCP,
+ NPC_SPORT_UDP,
+ NPC_DPORT_UDP,
+ NPC_SPORT_SCTP,
+ NPC_DPORT_SCTP,
+ NPC_HEADER_FIELDS_MAX,
+ NPC_CHAN = NPC_HEADER_FIELDS_MAX, /* Valid when Rx */
+ NPC_PF_FUNC, /* Valid when Tx */
+ NPC_ERRLEV,
+ NPC_ERRCODE,
+ NPC_LXMB,
+ NPC_LA,
+ NPC_LB,
+ NPC_LC,
+ NPC_LD,
+ NPC_LE,
+ NPC_LF,
+ NPC_LG,
+ NPC_LH,
+ /* ether type for untagged frame */
+ NPC_ETYPE_ETHER,
+ /* ether type for single tagged frame */
+ NPC_ETYPE_TAG1,
+ /* ether type for double tagged frame */
+ NPC_ETYPE_TAG2,
+ /* outer vlan tci for single tagged frame */
+ NPC_VLAN_TAG1,
+ /* outer vlan tci for double tagged frame */
+ NPC_VLAN_TAG2,
+ /* other header fields programmed to extract but not of our interest */
+ NPC_UNKNOWN,
+ NPC_KEY_FIELDS_MAX,
+};
+
struct npc_kpu_profile_cam {
u8 state;
u8 state_mask;
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
index 1724dbd18847..7e556c7b6ccf 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
@@ -15,6 +15,7 @@
#include "rvu_struct.h"
#include "common.h"
#include "mbox.h"
+#include "npc.h"
/* PCI device IDs */
#define PCI_DEVID_OCTEONTX2_RVU_AF 0xA065
@@ -105,6 +106,36 @@ struct nix_mce_list {
int max;
};
+/* layer meta data to uniquely identify a packet header field */
+struct npc_layer_mdata {
+ u8 lid;
+ u8 ltype;
+ u8 hdr;
+ u8 key;
+ u8 len;
+};
+
+/* Structure to represent a field present in the
+ * generated key. A key field may present anywhere and can
+ * be of any size in the generated key. Once this structure
+ * is populated for fields of interest then field's presence
+ * and location (if present) can be known.
+ */
+struct npc_key_field {
+ /* Masks where all set bits indicate position
+ * of a field in the key
+ */
+ u64 kw_mask[NPC_MAX_KWS_IN_KEY];
+ /* Number of words in the key a field spans. If a field is
+ * of 16 bytes and key offset is 4 then the field will use
+ * 4 bytes in KW0, 8 bytes in KW1 and 4 bytes in KW2 and
+ * nr_kws will be 3(KW0, KW1 and KW2).
+ */
+ int nr_kws;
+ /* used by packet header fields */
+ struct npc_layer_mdata layer_mdata;
+};
+
struct npc_mcam {
struct rsrc_bmap counters;
struct mutex lock; /* MCAM entries and counters update lock */
@@ -128,6 +159,11 @@ struct npc_mcam {
u16 hprio_count;
u16 hprio_end;
u16 rx_miss_act_cntr; /* Counter for RX MISS action */
+ /* fields present in the generated key */
+ struct npc_key_field tx_key_fields[NPC_KEY_FIELDS_MAX];
+ struct npc_key_field rx_key_fields[NPC_KEY_FIELDS_MAX];
+ u64 tx_features;
+ u64 rx_features;
};
/* Structure for per RVU func info ie PF/VF */
@@ -537,6 +573,8 @@ bool is_npc_intf_rx(u8 intf);
bool is_npc_interface_valid(struct rvu *rvu, u8 intf);
int rvu_npc_get_tx_nibble_cfg(struct rvu *rvu, u64 nibble_ena);
int npc_mcam_verify_channel(struct rvu *rvu, u16 pcifunc, u8 intf, u16 channel);
+int npc_flow_steering_init(struct rvu *rvu, int blkaddr);
+const char *npc_get_field_name(u8 hdr);
#ifdef CONFIG_DEBUG_FS
void rvu_dbg_init(struct rvu *rvu);
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
index 3666159bb6b6..eb4eaa7ece3a 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
@@ -1400,12 +1400,19 @@ int rvu_npc_init(struct rvu *rvu)
rvu_npc_setup_interfaces(rvu, blkaddr);
+ /* Configure MKEX profile */
+ npc_load_mkex_profile(rvu, blkaddr, rvu->mkex_pfl_name);
+
err = npc_mcam_rsrcs_init(rvu, blkaddr);
if (err)
return err;
- /* Configure MKEX profile */
- npc_load_mkex_profile(rvu, blkaddr, rvu->mkex_pfl_name);
+ err = npc_flow_steering_init(rvu, blkaddr);
+ if (err) {
+ dev_err(rvu->dev,
+ "Incorrect mkex profile loaded using default mkex\n");
+ npc_load_mkex_profile(rvu, blkaddr, def_pfl_name);
+ }
return 0;
}
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c
new file mode 100644
index 000000000000..4dae89776422
--- /dev/null
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c
@@ -0,0 +1,562 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Marvell OcteonTx2 RVU Admin Function driver
+ *
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#include <linux/bitfield.h>
+
+#include "rvu_struct.h"
+#include "rvu_reg.h"
+#include "rvu.h"
+#include "npc.h"
+
+#define NPC_BYTESM GENMASK_ULL(19, 16)
+#define NPC_HDR_OFFSET GENMASK_ULL(15, 8)
+#define NPC_KEY_OFFSET GENMASK_ULL(5, 0)
+#define NPC_LDATA_EN BIT_ULL(7)
+
+static const char * const npc_flow_names[] = {
+ [NPC_DMAC] = "dmac",
+ [NPC_SMAC] = "smac",
+ [NPC_ETYPE] = "ether type",
+ [NPC_OUTER_VID] = "outer vlan id",
+ [NPC_TOS] = "tos",
+ [NPC_SIP_IPV4] = "ipv4 source ip",
+ [NPC_DIP_IPV4] = "ipv4 destination ip",
+ [NPC_SIP_IPV6] = "ipv6 source ip",
+ [NPC_DIP_IPV6] = "ipv6 destination ip",
+ [NPC_SPORT_TCP] = "tcp source port",
+ [NPC_DPORT_TCP] = "tcp destination port",
+ [NPC_SPORT_UDP] = "udp source port",
+ [NPC_DPORT_UDP] = "udp destination port",
+ [NPC_SPORT_SCTP] = "sctp source port",
+ [NPC_DPORT_SCTP] = "sctp destination port",
+ [NPC_UNKNOWN] = "unknown",
+};
+
+const char *npc_get_field_name(u8 hdr)
+{
+ if (hdr >= ARRAY_SIZE(npc_flow_names))
+ return npc_flow_names[NPC_UNKNOWN];
+
+ return npc_flow_names[hdr];
+}
+
+/* Compute keyword masks and figure out the number of keywords a field
+ * spans in the key.
+ */
+static void npc_set_kw_masks(struct npc_mcam *mcam, u8 type,
+ u8 nr_bits, int start_kwi, int offset, u8 intf)
+{
+ struct npc_key_field *field = &mcam->rx_key_fields[type];
+ u8 bits_in_kw;
+ int max_kwi;
+
+ if (mcam->banks_per_entry == 1)
+ max_kwi = 1; /* NPC_MCAM_KEY_X1 */
+ else if (mcam->banks_per_entry == 2)
+ max_kwi = 3; /* NPC_MCAM_KEY_X2 */
+ else
+ max_kwi = 6; /* NPC_MCAM_KEY_X4 */
+
+ if (is_npc_intf_tx(intf))
+ field = &mcam->tx_key_fields[type];
+
+ if (offset + nr_bits <= 64) {
+ /* one KW only */
+ if (start_kwi > max_kwi)
+ return;
+ field->kw_mask[start_kwi] |= GENMASK_ULL(nr_bits - 1, 0)
+ << offset;
+ field->nr_kws = 1;
+ } else if (offset + nr_bits > 64 &&
+ offset + nr_bits <= 128) {
+ /* two KWs */
+ if (start_kwi + 1 > max_kwi)
+ return;
+ /* first KW mask */
+ bits_in_kw = 64 - offset;
+ field->kw_mask[start_kwi] |= GENMASK_ULL(bits_in_kw - 1, 0)
+ << offset;
+ /* second KW mask i.e. mask for rest of bits */
+ bits_in_kw = nr_bits + offset - 64;
+ field->kw_mask[start_kwi + 1] |= GENMASK_ULL(bits_in_kw - 1, 0);
+ field->nr_kws = 2;
+ } else {
+ /* three KWs */
+ if (start_kwi + 2 > max_kwi)
+ return;
+ /* first KW mask */
+ bits_in_kw = 64 - offset;
+ field->kw_mask[start_kwi] |= GENMASK_ULL(bits_in_kw - 1, 0)
+ << offset;
+ /* second KW mask */
+ field->kw_mask[start_kwi + 1] = ~0ULL;
+ /* third KW mask i.e. mask for rest of bits */
+ bits_in_kw = nr_bits + offset - 128;
+ field->kw_mask[start_kwi + 2] |= GENMASK_ULL(bits_in_kw - 1, 0);
+ field->nr_kws = 3;
+ }
+}
+
+/* Helper function to figure out whether field exists in the key */
+static bool npc_is_field_present(struct rvu *rvu, enum key_fields type, u8 intf)
+{
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ struct npc_key_field *input;
+
+ input = &mcam->rx_key_fields[type];
+ if (is_npc_intf_tx(intf))
+ input = &mcam->tx_key_fields[type];
+
+ return input->nr_kws > 0;
+}
+
+static bool npc_is_same(struct npc_key_field *input,
+ struct npc_key_field *field)
+{
+ int ret;
+
+ ret = memcmp(&input->layer_mdata, &field->layer_mdata,
+ sizeof(struct npc_layer_mdata));
+ return ret == 0;
+}
+
+static void npc_set_layer_mdata(struct npc_mcam *mcam, enum key_fields type,
+ u64 cfg, u8 lid, u8 lt, u8 intf)
+{
+ struct npc_key_field *input = &mcam->rx_key_fields[type];
+
+ if (is_npc_intf_tx(intf))
+ input = &mcam->tx_key_fields[type];
+
+ input->layer_mdata.hdr = FIELD_GET(NPC_HDR_OFFSET, cfg);
+ input->layer_mdata.key = FIELD_GET(NPC_KEY_OFFSET, cfg);
+ input->layer_mdata.len = FIELD_GET(NPC_BYTESM, cfg) + 1;
+ input->layer_mdata.ltype = lt;
+ input->layer_mdata.lid = lid;
+}
+
+static bool npc_check_overlap_fields(struct npc_key_field *input1,
+ struct npc_key_field *input2)
+{
+ int kwi;
+
+ /* Fields with same layer id and different ltypes are mutually
+ * exclusive hence they can be overlapped
+ */
+ if (input1->layer_mdata.lid == input2->layer_mdata.lid &&
+ input1->layer_mdata.ltype != input2->layer_mdata.ltype)
+ return false;
+
+ for (kwi = 0; kwi < NPC_MAX_KWS_IN_KEY; kwi++) {
+ if (input1->kw_mask[kwi] & input2->kw_mask[kwi])
+ return true;
+ }
+
+ return false;
+}
+
+/* Helper function to check whether given field overlaps with any other fields
+ * in the key. Due to limitations on key size and the key extraction profile in
+ * use higher layers can overwrite lower layer's header fields. Hence overlap
+ * needs to be checked.
+ */
+static bool npc_check_overlap(struct rvu *rvu, int blkaddr,
+ enum key_fields type, u8 start_lid, u8 intf)
+{
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ struct npc_key_field *dummy, *input;
+ int start_kwi, offset;
+ u8 nr_bits, lid, lt, ld;
+ u64 cfg;
+
+ dummy = &mcam->rx_key_fields[NPC_UNKNOWN];
+ input = &mcam->rx_key_fields[type];
+
+ if (is_npc_intf_tx(intf)) {
+ dummy = &mcam->tx_key_fields[NPC_UNKNOWN];
+ input = &mcam->tx_key_fields[type];
+ }
+
+ for (lid = start_lid; lid < NPC_MAX_LID; lid++) {
+ for (lt = 0; lt < NPC_MAX_LT; lt++) {
+ for (ld = 0; ld < NPC_MAX_LD; ld++) {
+ cfg = rvu_read64(rvu, blkaddr,
+ NPC_AF_INTFX_LIDX_LTX_LDX_CFG
+ (intf, lid, lt, ld));
+ if (!FIELD_GET(NPC_LDATA_EN, cfg))
+ continue;
+ memset(dummy, 0, sizeof(struct npc_key_field));
+ npc_set_layer_mdata(mcam, NPC_UNKNOWN, cfg,
+ lid, lt, intf);
+ /* exclude input */
+ if (npc_is_same(input, dummy))
+ continue;
+ start_kwi = dummy->layer_mdata.key / 8;
+ offset = (dummy->layer_mdata.key * 8) % 64;
+ nr_bits = dummy->layer_mdata.len * 8;
+ /* form KW masks */
+ npc_set_kw_masks(mcam, NPC_UNKNOWN, nr_bits,
+ start_kwi, offset, intf);
+ /* check any input field bits falls in any
+ * other field bits.
+ */
+ if (npc_check_overlap_fields(dummy, input))
+ return true;
+ }
+ }
+ }
+
+ return false;
+}
+
+static int npc_check_field(struct rvu *rvu, int blkaddr, enum key_fields type,
+ u8 intf)
+{
+ if (!npc_is_field_present(rvu, type, intf) ||
+ npc_check_overlap(rvu, blkaddr, type, 0, intf))
+ return -EOPNOTSUPP;
+ return 0;
+}
+
+static void npc_scan_parse_result(struct npc_mcam *mcam, u8 bit_number,
+ u8 key_nibble, u8 intf)
+{
+ u8 offset = (key_nibble * 4) % 64; /* offset within key word */
+ u8 kwi = (key_nibble * 4) / 64; /* which word in key */
+ u8 nr_bits = 4; /* bits in a nibble */
+ u8 type;
+
+ switch (bit_number) {
+ case 0 ... 2:
+ type = NPC_CHAN;
+ break;
+ case 3:
+ type = NPC_ERRLEV;
+ break;
+ case 4 ... 5:
+ type = NPC_ERRCODE;
+ break;
+ case 6:
+ type = NPC_LXMB;
+ break;
+ /* check for LTYPE only as of now */
+ case 9:
+ type = NPC_LA;
+ break;
+ case 12:
+ type = NPC_LB;
+ break;
+ case 15:
+ type = NPC_LC;
+ break;
+ case 18:
+ type = NPC_LD;
+ break;
+ case 21:
+ type = NPC_LE;
+ break;
+ case 24:
+ type = NPC_LF;
+ break;
+ case 27:
+ type = NPC_LG;
+ break;
+ case 30:
+ type = NPC_LH;
+ break;
+ default:
+ return;
+ };
+ npc_set_kw_masks(mcam, type, nr_bits, kwi, offset, intf);
+}
+
+static void npc_handle_multi_layer_fields(struct rvu *rvu, int blkaddr, u8 intf)
+{
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ struct npc_key_field *key_fields;
+ /* Ether type can come from three layers
+ * (ethernet, single tagged, double tagged)
+ */
+ struct npc_key_field *etype_ether;
+ struct npc_key_field *etype_tag1;
+ struct npc_key_field *etype_tag2;
+ /* Outer VLAN TCI can come from two layers
+ * (single tagged, double tagged)
+ */
+ struct npc_key_field *vlan_tag1;
+ struct npc_key_field *vlan_tag2;
+ u64 *features;
+ u8 start_lid;
+ int i;
+
+ key_fields = mcam->rx_key_fields;
+ features = &mcam->rx_features;
+
+ if (is_npc_intf_tx(intf)) {
+ key_fields = mcam->tx_key_fields;
+ features = &mcam->tx_features;
+ }
+
+ /* Handle header fields which can come from multiple layers like
+ * etype, outer vlan tci. These fields should have same position in
+ * the key otherwise to install a mcam rule more than one entry is
+ * needed which complicates mcam space management.
+ */
+ etype_ether = &key_fields[NPC_ETYPE_ETHER];
+ etype_tag1 = &key_fields[NPC_ETYPE_TAG1];
+ etype_tag2 = &key_fields[NPC_ETYPE_TAG2];
+ vlan_tag1 = &key_fields[NPC_VLAN_TAG1];
+ vlan_tag2 = &key_fields[NPC_VLAN_TAG2];
+
+ /* if key profile programmed does not extract ether type at all */
+ if (!etype_ether->nr_kws && !etype_tag1->nr_kws && !etype_tag2->nr_kws)
+ goto vlan_tci;
+
+ /* if key profile programmed extracts ether type from one layer */
+ if (etype_ether->nr_kws && !etype_tag1->nr_kws && !etype_tag2->nr_kws)
+ key_fields[NPC_ETYPE] = *etype_ether;
+ if (!etype_ether->nr_kws && etype_tag1->nr_kws && !etype_tag2->nr_kws)
+ key_fields[NPC_ETYPE] = *etype_tag1;
+ if (!etype_ether->nr_kws && !etype_tag1->nr_kws && etype_tag2->nr_kws)
+ key_fields[NPC_ETYPE] = *etype_tag2;
+
+ /* if key profile programmed extracts ether type from multiple layers */
+ if (etype_ether->nr_kws && etype_tag1->nr_kws) {
+ for (i = 0; i < NPC_MAX_KWS_IN_KEY; i++) {
+ if (etype_ether->kw_mask[i] != etype_tag1->kw_mask[i])
+ goto vlan_tci;
+ }
+ key_fields[NPC_ETYPE] = *etype_tag1;
+ }
+ if (etype_ether->nr_kws && etype_tag2->nr_kws) {
+ for (i = 0; i < NPC_MAX_KWS_IN_KEY; i++) {
+ if (etype_ether->kw_mask[i] != etype_tag2->kw_mask[i])
+ goto vlan_tci;
+ }
+ key_fields[NPC_ETYPE] = *etype_tag2;
+ }
+ if (etype_tag1->nr_kws && etype_tag2->nr_kws) {
+ for (i = 0; i < NPC_MAX_KWS_IN_KEY; i++) {
+ if (etype_tag1->kw_mask[i] != etype_tag2->kw_mask[i])
+ goto vlan_tci;
+ }
+ key_fields[NPC_ETYPE] = *etype_tag2;
+ }
+
+ /* check none of higher layers overwrite ether type */
+ start_lid = key_fields[NPC_ETYPE].layer_mdata.lid + 1;
+ if (npc_check_overlap(rvu, blkaddr, NPC_ETYPE, start_lid, intf))
+ goto vlan_tci;
+ *features |= BIT_ULL(NPC_ETYPE);
+vlan_tci:
+ /* if key profile does not extract outer vlan tci at all */
+ if (!vlan_tag1->nr_kws && !vlan_tag2->nr_kws)
+ goto done;
+
+ /* if key profile extracts outer vlan tci from one layer */
+ if (vlan_tag1->nr_kws && !vlan_tag2->nr_kws)
+ key_fields[NPC_OUTER_VID] = *vlan_tag1;
+ if (!vlan_tag1->nr_kws && vlan_tag2->nr_kws)
+ key_fields[NPC_OUTER_VID] = *vlan_tag2;
+
+ /* if key profile extracts outer vlan tci from multiple layers */
+ if (vlan_tag1->nr_kws && vlan_tag2->nr_kws) {
+ for (i = 0; i < NPC_MAX_KWS_IN_KEY; i++) {
+ if (vlan_tag1->kw_mask[i] != vlan_tag2->kw_mask[i])
+ goto done;
+ }
+ key_fields[NPC_OUTER_VID] = *vlan_tag2;
+ }
+ /* check none of higher layers overwrite outer vlan tci */
+ start_lid = key_fields[NPC_OUTER_VID].layer_mdata.lid + 1;
+ if (npc_check_overlap(rvu, blkaddr, NPC_OUTER_VID, start_lid, intf))
+ goto done;
+ *features |= BIT_ULL(NPC_OUTER_VID);
+done:
+ return;
+}
+
+static void npc_scan_ldata(struct rvu *rvu, int blkaddr, u8 lid,
+ u8 lt, u64 cfg, u8 intf)
+{
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ u8 hdr, key, nr_bytes, bit_offset;
+ u8 la_ltype, la_start;
+ /* starting KW index and starting bit position */
+ int start_kwi, offset;
+
+ nr_bytes = FIELD_GET(NPC_BYTESM, cfg) + 1;
+ hdr = FIELD_GET(NPC_HDR_OFFSET, cfg);
+ key = FIELD_GET(NPC_KEY_OFFSET, cfg);
+ start_kwi = key / 8;
+ offset = (key * 8) % 64;
+
+ /* For Tx, Layer A has NIX_INST_HDR_S(64 bytes) preceding
+ * ethernet header.
+ */
+ if (is_npc_intf_tx(intf)) {
+ la_ltype = NPC_LT_LA_IH_NIX_ETHER;
+ la_start = 8;
+ } else {
+ la_ltype = NPC_LT_LA_ETHER;
+ la_start = 0;
+ }
+
+#define NPC_SCAN_HDR(name, hlid, hlt, hstart, hlen) \
+do { \
+ if (lid == (hlid) && lt == (hlt)) { \
+ if ((hstart) >= hdr && \
+ ((hstart) + (hlen)) <= (hdr + nr_bytes)) { \
+ bit_offset = (hdr + nr_bytes - (hstart) - (hlen)) * 8; \
+ npc_set_layer_mdata(mcam, (name), cfg, lid, lt, intf); \
+ npc_set_kw_masks(mcam, (name), (hlen) * 8, \
+ start_kwi, offset + bit_offset, intf);\
+ } \
+ } \
+} while (0)
+
+ /* List LID, LTYPE, start offset from layer and length(in bytes) of
+ * packet header fields below.
+ * Example: Source IP is 4 bytes and starts at 12th byte of IP header
+ */
+ NPC_SCAN_HDR(NPC_SIP_IPV4, NPC_LID_LC, NPC_LT_LC_IP, 12, 4);
+ NPC_SCAN_HDR(NPC_DIP_IPV4, NPC_LID_LC, NPC_LT_LC_IP, 16, 4);
+ NPC_SCAN_HDR(NPC_SIP_IPV6, NPC_LID_LC, NPC_LT_LC_IP6, 8, 16);
+ NPC_SCAN_HDR(NPC_DIP_IPV6, NPC_LID_LC, NPC_LT_LC_IP6, 24, 16);
+ NPC_SCAN_HDR(NPC_SPORT_UDP, NPC_LID_LD, NPC_LT_LD_UDP, 0, 2);
+ NPC_SCAN_HDR(NPC_DPORT_UDP, NPC_LID_LD, NPC_LT_LD_UDP, 2, 2);
+ NPC_SCAN_HDR(NPC_SPORT_TCP, NPC_LID_LD, NPC_LT_LD_TCP, 0, 2);
+ NPC_SCAN_HDR(NPC_DPORT_TCP, NPC_LID_LD, NPC_LT_LD_TCP, 2, 2);
+ NPC_SCAN_HDR(NPC_SPORT_SCTP, NPC_LID_LD, NPC_LT_LD_SCTP, 0, 2);
+ NPC_SCAN_HDR(NPC_DPORT_SCTP, NPC_LID_LD, NPC_LT_LD_SCTP, 2, 2);
+ NPC_SCAN_HDR(NPC_ETYPE_ETHER, NPC_LID_LA, NPC_LT_LA_ETHER, 12, 2);
+ NPC_SCAN_HDR(NPC_ETYPE_TAG1, NPC_LID_LB, NPC_LT_LB_CTAG, 4, 2);
+ NPC_SCAN_HDR(NPC_ETYPE_TAG2, NPC_LID_LB, NPC_LT_LB_STAG_QINQ, 8, 2);
+ NPC_SCAN_HDR(NPC_VLAN_TAG1, NPC_LID_LB, NPC_LT_LB_CTAG, 2, 2);
+ NPC_SCAN_HDR(NPC_VLAN_TAG2, NPC_LID_LB, NPC_LT_LB_STAG_QINQ, 2, 2);
+ NPC_SCAN_HDR(NPC_DMAC, NPC_LID_LA, la_ltype, la_start, 6);
+ NPC_SCAN_HDR(NPC_SMAC, NPC_LID_LA, la_ltype, la_start, 6);
+ /* PF_FUNC is 2 bytes at 0th byte of NPC_LT_LA_IH_NIX_ETHER */
+ NPC_SCAN_HDR(NPC_PF_FUNC, NPC_LID_LA, NPC_LT_LA_IH_NIX_ETHER, 0, 2);
+}
+
+static void npc_set_features(struct rvu *rvu, int blkaddr, u8 intf)
+{
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ u64 *features = &mcam->rx_features;
+ u64 tcp_udp_sctp;
+ int err, hdr;
+
+ if (is_npc_intf_tx(intf))
+ features = &mcam->tx_features;
+
+ for (hdr = NPC_DMAC; hdr < NPC_HEADER_FIELDS_MAX; hdr++) {
+ err = npc_check_field(rvu, blkaddr, hdr, intf);
+ if (!err)
+ *features |= BIT_ULL(hdr);
+ }
+
+ tcp_udp_sctp = BIT_ULL(NPC_SPORT_TCP) | BIT_ULL(NPC_SPORT_UDP) |
+ BIT_ULL(NPC_DPORT_TCP) | BIT_ULL(NPC_DPORT_UDP) |
+ BIT_ULL(NPC_SPORT_SCTP) | BIT_ULL(NPC_DPORT_SCTP);
+
+ /* for tcp/udp/sctp corresponding layer type should be in the key */
+ if (*features & tcp_udp_sctp)
+ if (npc_check_field(rvu, blkaddr, NPC_LD, intf))
+ *features &= ~tcp_udp_sctp;
+
+ /* for vlan corresponding layer type should be in the key */
+ if (*features & BIT_ULL(NPC_OUTER_VID))
+ if (npc_check_field(rvu, blkaddr, NPC_LB, intf))
+ *features &= ~BIT_ULL(NPC_OUTER_VID);
+}
+
+/* Scan key extraction profile and record how fields of our interest
+ * fill the key structure. Also verify Channel and DMAC exists in
+ * key and not overwritten by other header fields.
+ */
+static int npc_scan_kex(struct rvu *rvu, int blkaddr, u8 intf)
+{
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ u8 lid, lt, ld, bitnr;
+ u8 key_nibble = 0;
+ u64 cfg;
+
+ /* Scan and note how parse result is going to be in key.
+ * A bit set in PARSE_NIBBLE_ENA corresponds to a nibble from
+ * parse result in the key. The enabled nibbles from parse result
+ * will be concatenated in key.
+ */
+ cfg = rvu_read64(rvu, blkaddr, NPC_AF_INTFX_KEX_CFG(intf));
+ cfg &= NPC_PARSE_NIBBLE;
+ for_each_set_bit(bitnr, (unsigned long *)&cfg, 31) {
+ npc_scan_parse_result(mcam, bitnr, key_nibble, intf);
+ key_nibble++;
+ }
+
+ /* Scan and note how layer data is going to be in key */
+ for (lid = 0; lid < NPC_MAX_LID; lid++) {
+ for (lt = 0; lt < NPC_MAX_LT; lt++) {
+ for (ld = 0; ld < NPC_MAX_LD; ld++) {
+ cfg = rvu_read64(rvu, blkaddr,
+ NPC_AF_INTFX_LIDX_LTX_LDX_CFG
+ (intf, lid, lt, ld));
+ if (!FIELD_GET(NPC_LDATA_EN, cfg))
+ continue;
+ npc_scan_ldata(rvu, blkaddr, lid, lt, cfg,
+ intf);
+ }
+ }
+ }
+
+ return 0;
+}
+
+static int npc_scan_verify_kex(struct rvu *rvu, int blkaddr)
+{
+ int err;
+
+ err = npc_scan_kex(rvu, blkaddr, NIX_INTF_RX);
+ if (err)
+ return err;
+
+ err = npc_scan_kex(rvu, blkaddr, NIX_INTF_TX);
+ if (err)
+ return err;
+
+ /* Channel is mandatory */
+ if (!npc_is_field_present(rvu, NPC_CHAN, NIX_INTF_RX)) {
+ dev_err(rvu->dev, "Channel not present in Key\n");
+ return -EINVAL;
+ }
+ /* check that none of the fields overwrite channel */
+ if (npc_check_overlap(rvu, blkaddr, NPC_CHAN, 0, NIX_INTF_RX)) {
+ dev_err(rvu->dev, "Channel cannot be overwritten\n");
+ return -EINVAL;
+ }
+ /* DMAC should be present in key for unicast filter to work */
+ if (!npc_is_field_present(rvu, NPC_DMAC, NIX_INTF_RX)) {
+ dev_err(rvu->dev, "DMAC not present in Key\n");
+ return -EINVAL;
+ }
+ /* check that none of the fields overwrite DMAC */
+ if (npc_check_overlap(rvu, blkaddr, NPC_DMAC, 0, NIX_INTF_RX)) {
+ dev_err(rvu->dev, "DMAC cannot be overwritten\n");
+ return -EINVAL;
+ }
+
+ npc_set_features(rvu, blkaddr, NIX_INTF_TX);
+ npc_set_features(rvu, blkaddr, NIX_INTF_RX);
+ npc_handle_multi_layer_fields(rvu, blkaddr, NIX_INTF_TX);
+ npc_handle_multi_layer_fields(rvu, blkaddr, NIX_INTF_RX);
+
+ return 0;
+}
+
+int npc_flow_steering_init(struct rvu *rvu, int blkaddr)
+{
+ return npc_scan_verify_kex(rvu, blkaddr);
+}
--
2.16.5
^ permalink raw reply related [flat|nested] 24+ messages in thread* Re: [PATCH v3 net-next 03/13] octeontx2-af: Generate key field bit mask from KEX profile
2020-11-11 7:13 ` [PATCH v3 net-next 03/13] octeontx2-af: Generate key field bit mask from KEX profile Naveen Mamindlapalli
@ 2020-11-12 21:24 ` Alexander Duyck
0 siblings, 0 replies; 24+ messages in thread
From: Alexander Duyck @ 2020-11-12 21:24 UTC (permalink / raw)
To: Naveen Mamindlapalli
Cc: Netdev, LKML, Jakub Kicinski, David Miller, saeed, sgoutham,
lcherian, gakula, jerinj, sbhatta, hkelam
On Tue, Nov 10, 2020 at 11:24 PM Naveen Mamindlapalli
<naveenm@marvell.com> wrote:
>
> From: Subbaraya Sundeep <sbhatta@marvell.com>
>
> Key Extraction(KEX) profile decides how the packet metadata such as
> layer information and selected packet data bytes at each layer are
> placed in MCAM search key. This patch reads the configured KEX profile
> parameters to find out the bit position and bit mask for each field.
> The information is used when programming the MCAM match data by SW
> to match a packet flow and take appropriate action on the flow. This
> patch also verifies the mandatory fields such as channel and DMAC
> are not overwritten by the KEX configuration of other fields.
>
> Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
> Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
> Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
A few minor spelling issues, otherwise it looks fine.
Reviewed-by: Alexander Duyck <alexanderduyck@fb.com>
> ---
> drivers/net/ethernet/marvell/octeontx2/af/Makefile | 2 +-
> drivers/net/ethernet/marvell/octeontx2/af/npc.h | 48 ++
> drivers/net/ethernet/marvell/octeontx2/af/rvu.h | 38 ++
> .../net/ethernet/marvell/octeontx2/af/rvu_npc.c | 11 +-
> .../net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c | 562 +++++++++++++++++++++
> 5 files changed, 658 insertions(+), 3 deletions(-)
> create mode 100644 drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c
>
> diff --git a/drivers/net/ethernet/marvell/octeontx2/af/Makefile b/drivers/net/ethernet/marvell/octeontx2/af/Makefile
> index 2f7a861d0c7b..ffc681b67f1c 100644
> --- a/drivers/net/ethernet/marvell/octeontx2/af/Makefile
> +++ b/drivers/net/ethernet/marvell/octeontx2/af/Makefile
> @@ -9,4 +9,4 @@ obj-$(CONFIG_OCTEONTX2_AF) += octeontx2_af.o
>
> octeontx2_mbox-y := mbox.o rvu_trace.o
> octeontx2_af-y := cgx.o rvu.o rvu_cgx.o rvu_npa.o rvu_nix.o \
> - rvu_reg.o rvu_npc.o rvu_debugfs.o ptp.o
> + rvu_reg.o rvu_npc.o rvu_debugfs.o ptp.o rvu_npc_fs.o
> diff --git a/drivers/net/ethernet/marvell/octeontx2/af/npc.h b/drivers/net/ethernet/marvell/octeontx2/af/npc.h
> index 91a9d00e4fb5..0fe47216f771 100644
> --- a/drivers/net/ethernet/marvell/octeontx2/af/npc.h
> +++ b/drivers/net/ethernet/marvell/octeontx2/af/npc.h
> @@ -140,6 +140,54 @@ enum npc_kpu_lh_ltype {
> NPC_LT_LH_CUSTOM1 = 0xF,
> };
>
> +/* list of known and supported fields in packet header and
> + * fields present in key structure.
> + */
> +enum key_fields {
> + NPC_DMAC,
> + NPC_SMAC,
> + NPC_ETYPE,
> + NPC_OUTER_VID,
> + NPC_TOS,
> + NPC_SIP_IPV4,
> + NPC_DIP_IPV4,
> + NPC_SIP_IPV6,
> + NPC_DIP_IPV6,
> + NPC_SPORT_TCP,
> + NPC_DPORT_TCP,
> + NPC_SPORT_UDP,
> + NPC_DPORT_UDP,
> + NPC_SPORT_SCTP,
> + NPC_DPORT_SCTP,
> + NPC_HEADER_FIELDS_MAX,
> + NPC_CHAN = NPC_HEADER_FIELDS_MAX, /* Valid when Rx */
> + NPC_PF_FUNC, /* Valid when Tx */
> + NPC_ERRLEV,
> + NPC_ERRCODE,
> + NPC_LXMB,
> + NPC_LA,
> + NPC_LB,
> + NPC_LC,
> + NPC_LD,
> + NPC_LE,
> + NPC_LF,
> + NPC_LG,
> + NPC_LH,
> + /* ether type for untagged frame */
> + NPC_ETYPE_ETHER,
> + /* ether type for single tagged frame */
> + NPC_ETYPE_TAG1,
> + /* ether type for double tagged frame */
> + NPC_ETYPE_TAG2,
> + /* outer vlan tci for single tagged frame */
> + NPC_VLAN_TAG1,
> + /* outer vlan tci for double tagged frame */
> + NPC_VLAN_TAG2,
> + /* other header fields programmed to extract but not of our interest */
> + NPC_UNKNOWN,
> + NPC_KEY_FIELDS_MAX,
> +};
> +
> struct npc_kpu_profile_cam {
> u8 state;
> u8 state_mask;
> diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
> index 1724dbd18847..7e556c7b6ccf 100644
> --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
> +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
> @@ -15,6 +15,7 @@
> #include "rvu_struct.h"
> #include "common.h"
> #include "mbox.h"
> +#include "npc.h"
>
> /* PCI device IDs */
> #define PCI_DEVID_OCTEONTX2_RVU_AF 0xA065
> @@ -105,6 +106,36 @@ struct nix_mce_list {
> int max;
> };
>
> +/* layer meta data to uniquely identify a packet header field */
s/meta data/metadata/
> +struct npc_layer_mdata {
> + u8 lid;
> + u8 ltype;
> + u8 hdr;
> + u8 key;
> + u8 len;
> +};
> +
<snip>
> + /* Handle header fields which can come from multiple layers like
> + * etype, outer vlan tci. These fields should have same position in
> + * the key otherwise to install a mcam rule more than one entry is
> + * needed which complicates mcam space management.
> + */
> + etype_ether = &key_fields[NPC_ETYPE_ETHER];
> + etype_tag1 = &key_fields[NPC_ETYPE_TAG1];
> + etype_tag2 = &key_fields[NPC_ETYPE_TAG2];
> + vlan_tag1 = &key_fields[NPC_VLAN_TAG1];
> + vlan_tag2 = &key_fields[NPC_VLAN_TAG2];
> +
> + /* if key profile programmed does not extract ether type at all */
s/ether type/Ethertype/
> + if (!etype_ether->nr_kws && !etype_tag1->nr_kws && !etype_tag2->nr_kws)
> + goto vlan_tci;
> +
> + /* if key profile programmed extracts ether type from one layer */
Same issue here and a few other places, replace "ether type" with "Ethertype".
> + if (etype_ether->nr_kws && !etype_tag1->nr_kws && !etype_tag2->nr_kws)
> + key_fields[NPC_ETYPE] = *etype_ether;
> + if (!etype_ether->nr_kws && etype_tag1->nr_kws && !etype_tag2->nr_kws)
> + key_fields[NPC_ETYPE] = *etype_tag1;
> + if (!etype_ether->nr_kws && !etype_tag1->nr_kw
>
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH v3 net-next 04/13] octeontx2-af: Add mbox messages to install and delete MCAM rules
2020-11-11 7:13 [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Naveen Mamindlapalli
` (2 preceding siblings ...)
2020-11-11 7:13 ` [PATCH v3 net-next 03/13] octeontx2-af: Generate key field bit mask from KEX profile Naveen Mamindlapalli
@ 2020-11-11 7:13 ` Naveen Mamindlapalli
2020-11-12 21:46 ` Alexander Duyck
2020-11-11 7:13 ` [PATCH v3 net-next 05/13] octeontx2-pf: Add support for ethtool ntuple filters Naveen Mamindlapalli
` (9 subsequent siblings)
13 siblings, 1 reply; 24+ messages in thread
From: Naveen Mamindlapalli @ 2020-11-11 7:13 UTC (permalink / raw)
To: netdev, linux-kernel
Cc: kuba, davem, saeed, sgoutham, lcherian, gakula, jerinj, sbhatta,
hkelam
From: Subbaraya Sundeep <sbhatta@marvell.com>
Added new mailbox messages to install and delete MCAM rules.
These mailbox messages will be used for adding/deleting ethtool
n-tuple filters by NIX PF. The installed MCAM rules are stored
in a list that will be traversed later to delete the MCAM entries
when the interface is brought down or when PCIe FLR is received.
The delete mailbox supports deleting a single MCAM entry or range
of entries or all the MCAM entries owned by the pcifunc. Each MCAM
entry can be associated with a HW match stat entry if the mailbox
requester wants to check the hit count for debugging.
Modified adding default unicast DMAC match rule using install
flow API. The default unicast DMAC match entry installed by
Administrative Function is saved and can be changed later by the
mailbox user to fit additional fields, or the default MCAM entry
rule action can be used for other flow rules installed later.
Modified rvu_mbox_handler_nix_lf_free mailbox to add a flag to
disable or delete the MCAM entries. The MCAM entries are disabled
when the interface is brought down and deleted in FLR handler.
The disabled MCAM entries will be re-enabled when the interface
is brought up again.
Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
---
drivers/net/ethernet/marvell/octeontx2/af/common.h | 2 +
drivers/net/ethernet/marvell/octeontx2/af/mbox.h | 76 ++-
drivers/net/ethernet/marvell/octeontx2/af/npc.h | 57 +-
drivers/net/ethernet/marvell/octeontx2/af/rvu.h | 13 +
.../net/ethernet/marvell/octeontx2/af/rvu_nix.c | 19 +-
.../net/ethernet/marvell/octeontx2/af/rvu_npc.c | 217 ++++++-
.../net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c | 721 +++++++++++++++++++++
.../net/ethernet/marvell/octeontx2/nic/otx2_pf.c | 12 +-
8 files changed, 1065 insertions(+), 52 deletions(-)
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/common.h b/drivers/net/ethernet/marvell/octeontx2/af/common.h
index 8f68e7a8b882..17f6f42f4453 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/common.h
@@ -162,6 +162,8 @@ enum nix_scheduler {
#define NIX_RX_ACTIONOP_UCAST_IPSEC (0x2ull)
#define NIX_RX_ACTIONOP_MCAST (0x3ull)
#define NIX_RX_ACTIONOP_RSS (0x4ull)
+/* Use the RX action set in the default unicast entry */
+#define NIX_RX_ACTION_DEFAULT (0xfull)
/* NIX TX action operation*/
#define NIX_TX_ACTIONOP_DROP (0x0ull)
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
index f46de8419b77..ac3118d2d126 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
@@ -188,10 +188,14 @@ M(NPC_MCAM_ALLOC_AND_WRITE_ENTRY, 0x600b, npc_mcam_alloc_and_write_entry, \
npc_mcam_alloc_and_write_entry_rsp) \
M(NPC_GET_KEX_CFG, 0x600c, npc_get_kex_cfg, \
msg_req, npc_get_kex_cfg_rsp) \
+M(NPC_INSTALL_FLOW, 0x600d, npc_install_flow, \
+ npc_install_flow_req, npc_install_flow_rsp) \
+M(NPC_DELETE_FLOW, 0x600e, npc_delete_flow, \
+ npc_delete_flow_req, msg_rsp) \
/* NIX mbox IDs (range 0x8000 - 0xFFFF) */ \
M(NIX_LF_ALLOC, 0x8000, nix_lf_alloc, \
nix_lf_alloc_req, nix_lf_alloc_rsp) \
-M(NIX_LF_FREE, 0x8001, nix_lf_free, msg_req, msg_rsp) \
+M(NIX_LF_FREE, 0x8001, nix_lf_free, nix_lf_free_req, msg_rsp) \
M(NIX_AQ_ENQ, 0x8002, nix_aq_enq, nix_aq_enq_req, nix_aq_enq_rsp) \
M(NIX_HWCTX_DISABLE, 0x8003, nix_hwctx_disable, \
hwctx_disable_req, msg_rsp) \
@@ -510,6 +514,12 @@ struct nix_lf_alloc_rsp {
u8 sdp_links; /* No. of SDP links present in HW */
};
+struct nix_lf_free_req {
+ struct mbox_msghdr hdr;
+#define NIX_LF_DISABLE_FLOWS BIT_ULL(0)
+ u64 flags;
+};
+
/* NIX AQ enqueue msg */
struct nix_aq_enq_req {
struct mbox_msghdr hdr;
@@ -882,6 +892,70 @@ struct npc_get_kex_cfg_rsp {
u8 mkex_pfl_name[MKEX_NAME_LEN];
};
+struct flow_msg {
+ unsigned char dmac[6];
+ unsigned char smac[6];
+ __be16 etype;
+ __be16 vlan_etype;
+ __be16 vlan_tci;
+ union {
+ __be32 ip4src;
+ __be32 ip6src[4];
+ };
+ union {
+ __be32 ip4dst;
+ __be32 ip6dst[4];
+ };
+ u8 tos;
+ u8 ip_ver;
+ u8 ip_proto;
+ u8 tc;
+ __be16 sport;
+ __be16 dport;
+};
+
+struct npc_install_flow_req {
+ struct mbox_msghdr hdr;
+ struct flow_msg packet;
+ struct flow_msg mask;
+ u64 features;
+ u16 entry;
+ u16 channel;
+ u8 intf;
+ u8 set_cntr; /* If counter is available set counter for this entry ? */
+ u8 default_rule;
+ u8 append; /* overwrite(0) or append(1) flow to default rule? */
+ u16 vf;
+ /* action */
+ u32 index;
+ u16 match_id;
+ u8 flow_key_alg;
+ u8 op;
+ /* vtag rx action */
+ u8 vtag0_type;
+ u8 vtag0_valid;
+ u8 vtag1_type;
+ u8 vtag1_valid;
+ /* vtag tx action */
+ u16 vtag0_def;
+ u8 vtag0_op;
+ u16 vtag1_def;
+ u8 vtag1_op;
+};
+
+struct npc_install_flow_rsp {
+ struct mbox_msghdr hdr;
+ int counter; /* negative if no counter else counter number */
+};
+
+struct npc_delete_flow_req {
+ struct mbox_msghdr hdr;
+ u16 entry;
+ u16 start;/*Disable range of entries */
+ u16 end;
+ u8 all; /* PF + VFs */
+};
+
enum ptp_op {
PTP_OP_ADJFINE = 0,
PTP_OP_GET_CLOCK = 1,
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/npc.h b/drivers/net/ethernet/marvell/octeontx2/af/npc.h
index 0fe47216f771..860706033aad 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/npc.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/npc.h
@@ -348,11 +348,41 @@ struct nix_rx_action {
/* NPC_AF_INTFX_KEX_CFG field masks */
#define NPC_PARSE_NIBBLE GENMASK_ULL(30, 0)
+struct nix_tx_action {
+#if defined(__BIG_ENDIAN_BITFIELD)
+ u64 rsvd_63_48 :16;
+ u64 match_id :16;
+ u64 index :20;
+ u64 rsvd_11_8 :8;
+ u64 op :4;
+#else
+ u64 op :4;
+ u64 rsvd_11_8 :8;
+ u64 index :20;
+ u64 match_id :16;
+ u64 rsvd_63_48 :16;
+#endif
+};
+
/* NIX Receive Vtag Action Structure */
-#define VTAG0_VALID_BIT BIT_ULL(15)
-#define VTAG0_TYPE_MASK GENMASK_ULL(14, 12)
-#define VTAG0_LID_MASK GENMASK_ULL(10, 8)
-#define VTAG0_RELPTR_MASK GENMASK_ULL(7, 0)
+#define RX_VTAG0_VALID_BIT BIT_ULL(15)
+#define RX_VTAG0_TYPE_MASK GENMASK_ULL(14, 12)
+#define RX_VTAG0_LID_MASK GENMASK_ULL(10, 8)
+#define RX_VTAG0_RELPTR_MASK GENMASK_ULL(7, 0)
+#define RX_VTAG1_VALID_BIT BIT_ULL(47)
+#define RX_VTAG1_TYPE_MASK GENMASK_ULL(46, 44)
+#define RX_VTAG1_LID_MASK GENMASK_ULL(42, 40)
+#define RX_VTAG1_RELPTR_MASK GENMASK_ULL(39, 32)
+
+/* NIX Transmit Vtag Action Structure */
+#define TX_VTAG0_DEF_MASK GENMASK_ULL(25, 16)
+#define TX_VTAG0_OP_MASK GENMASK_ULL(13, 12)
+#define TX_VTAG0_LID_MASK GENMASK_ULL(10, 8)
+#define TX_VTAG0_RELPTR_MASK GENMASK_ULL(7, 0)
+#define TX_VTAG1_DEF_MASK GENMASK_ULL(57, 48)
+#define TX_VTAG1_OP_MASK GENMASK_ULL(45, 44)
+#define TX_VTAG1_LID_MASK GENMASK_ULL(42, 40)
+#define TX_VTAG1_RELPTR_MASK GENMASK_ULL(39, 32)
struct npc_mcam_kex {
/* MKEX Profle Header */
@@ -405,4 +435,23 @@ struct npc_lt_def_cfg {
struct npc_lt_def pck_iip4;
};
+struct rvu_npc_mcam_rule {
+ struct flow_msg packet;
+ struct flow_msg mask;
+ u8 intf;
+ union {
+ struct nix_tx_action tx_action;
+ struct nix_rx_action rx_action;
+ };
+ u64 vtag_action;
+ struct list_head list;
+ u64 features;
+ u16 owner;
+ u16 entry;
+ u16 cntr;
+ bool has_cntr;
+ u8 default_rule;
+ bool enable;
+};
+
#endif /* NPC_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
index 7e556c7b6ccf..fecde9e73a33 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
@@ -147,6 +147,7 @@ struct npc_mcam {
u16 *entry2cntr_map;
u16 *cntr2pfvf_map;
u16 *cntr_refcnt;
+ u16 *entry2target_pffunc;
u8 keysize; /* MCAM keysize 112/224/448 bits */
u8 banks; /* Number of MCAM banks */
u8 banks_per_entry;/* Number of keywords in key */
@@ -164,6 +165,7 @@ struct npc_mcam {
struct npc_key_field rx_key_fields[NPC_KEY_FIELDS_MAX];
u64 tx_features;
u64 rx_features;
+ struct list_head mcam_rules;
};
/* Structure for per RVU func info ie PF/VF */
@@ -218,6 +220,8 @@ struct rvu_pfvf {
int rxvlan_index;
bool rxvlan;
+ struct rvu_npc_mcam_rule *def_ucast_rule;
+
bool cgx_in_use; /* this PF/VF using CGX? */
int cgx_users; /* number of cgx users - used only by PFs */
@@ -558,6 +562,7 @@ void rvu_npc_install_bcast_match_entry(struct rvu *rvu, u16 pcifunc,
void rvu_npc_enable_bcast_entry(struct rvu *rvu, u16 pcifunc, bool enable);
int rvu_npc_update_rxvlan(struct rvu *rvu, u16 pcifunc, int nixlf);
void rvu_npc_disable_mcam_entries(struct rvu *rvu, u16 pcifunc, int nixlf);
+void rvu_npc_free_mcam_entries(struct rvu *rvu, u16 pcifunc, int nixlf);
void rvu_npc_disable_default_entries(struct rvu *rvu, u16 pcifunc, int nixlf);
void rvu_npc_enable_default_entries(struct rvu *rvu, u16 pcifunc, int nixlf);
void rvu_npc_update_flowkey_alg_idx(struct rvu *rvu, u16 pcifunc, int nixlf,
@@ -575,6 +580,14 @@ int rvu_npc_get_tx_nibble_cfg(struct rvu *rvu, u64 nibble_ena);
int npc_mcam_verify_channel(struct rvu *rvu, u16 pcifunc, u8 intf, u16 channel);
int npc_flow_steering_init(struct rvu *rvu, int blkaddr);
const char *npc_get_field_name(u8 hdr);
+bool rvu_npc_write_default_rule(struct rvu *rvu, int blkaddr, int nixlf,
+ u16 pcifunc, u8 intf, struct mcam_entry *entry,
+ int *entry_index);
+int npc_get_bank(struct npc_mcam *mcam, int index);
+void npc_mcam_enable_flows(struct rvu *rvu, u16 target);
+void npc_mcam_disable_flows(struct rvu *rvu, u16 target);
+void npc_enable_mcam_entry(struct rvu *rvu, struct npc_mcam *mcam,
+ int blkaddr, int index, bool enable);
#ifdef CONFIG_DEBUG_FS
void rvu_dbg_init(struct rvu *rvu);
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
index 8c11abdbd9d1..9314309da539 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
@@ -304,7 +304,6 @@ static void nix_interface_deinit(struct rvu *rvu, u16 pcifunc, u8 nixlf)
pfvf->maxlen = 0;
pfvf->minlen = 0;
- pfvf->rxvlan = false;
/* Remove this PF_FUNC from bcast pkt replication list */
err = nix_update_bcast_mce_list(rvu, pcifunc, false);
@@ -1230,7 +1229,7 @@ int rvu_mbox_handler_nix_lf_alloc(struct rvu *rvu,
return rc;
}
-int rvu_mbox_handler_nix_lf_free(struct rvu *rvu, struct msg_req *req,
+int rvu_mbox_handler_nix_lf_free(struct rvu *rvu, struct nix_lf_free_req *req,
struct msg_rsp *rsp)
{
struct rvu_hwinfo *hw = rvu->hw;
@@ -1249,6 +1248,11 @@ int rvu_mbox_handler_nix_lf_free(struct rvu *rvu, struct msg_req *req,
if (nixlf < 0)
return NIX_AF_ERR_AF_LF_INVALID;
+ if (req->flags & NIX_LF_DISABLE_FLOWS)
+ rvu_npc_disable_mcam_entries(rvu, pcifunc, nixlf);
+ else
+ rvu_npc_free_mcam_entries(rvu, pcifunc, nixlf);
+
nix_interface_deinit(rvu, pcifunc, nixlf);
/* Reset this NIX LF */
@@ -2764,8 +2768,6 @@ int rvu_mbox_handler_nix_set_mac_addr(struct rvu *rvu,
rvu_npc_install_ucast_entry(rvu, pcifunc, nixlf,
pfvf->rx_chan_base, req->mac_addr);
- rvu_npc_update_rxvlan(rvu, pcifunc, nixlf);
-
return 0;
}
@@ -2812,9 +2814,6 @@ int rvu_mbox_handler_nix_set_rx_mode(struct rvu *rvu, struct nix_rx_mode *req,
else
rvu_npc_install_promisc_entry(rvu, pcifunc, nixlf,
pfvf->rx_chan_base, allmulti);
-
- rvu_npc_update_rxvlan(rvu, pcifunc, nixlf);
-
return 0;
}
@@ -3378,6 +3377,8 @@ int rvu_mbox_handler_nix_lf_start_rx(struct rvu *rvu, struct msg_req *req,
rvu_npc_enable_default_entries(rvu, pcifunc, nixlf);
+ npc_mcam_enable_flows(rvu, pcifunc);
+
return rvu_cgx_start_stop_io(rvu, pcifunc, true);
}
@@ -3393,6 +3394,8 @@ int rvu_mbox_handler_nix_lf_stop_rx(struct rvu *rvu, struct msg_req *req,
rvu_npc_disable_default_entries(rvu, pcifunc, nixlf);
+ npc_mcam_disable_flows(rvu, pcifunc);
+
return rvu_cgx_start_stop_io(rvu, pcifunc, false);
}
@@ -3405,6 +3408,8 @@ void rvu_nix_lf_teardown(struct rvu *rvu, u16 pcifunc, int blkaddr, int nixlf)
ctx_req.hdr.pcifunc = pcifunc;
/* Cleanup NPC MCAM entries, free Tx scheduler queues being used */
+ rvu_npc_disable_mcam_entries(rvu, pcifunc, nixlf);
+ rvu_npc_free_mcam_entries(rvu, pcifunc, nixlf);
nix_interface_deinit(rvu, pcifunc, nixlf);
nix_rx_sync(rvu, blkaddr);
nix_txschq_free(rvu, pcifunc);
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
index eb4eaa7ece3a..a7759ecfa586 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
@@ -219,7 +219,7 @@ static int npc_get_nixlf_mcam_index(struct npc_mcam *mcam,
return npc_get_ucast_mcam_index(mcam, pcifunc, nixlf);
}
-static int npc_get_bank(struct npc_mcam *mcam, int index)
+int npc_get_bank(struct npc_mcam *mcam, int index)
{
int bank = index / mcam->banksize;
@@ -241,8 +241,8 @@ static bool is_mcam_entry_enabled(struct rvu *rvu, struct npc_mcam *mcam,
return (cfg & 1);
}
-static void npc_enable_mcam_entry(struct rvu *rvu, struct npc_mcam *mcam,
- int blkaddr, int index, bool enable)
+void npc_enable_mcam_entry(struct rvu *rvu, struct npc_mcam *mcam,
+ int blkaddr, int index, bool enable)
{
int bank = npc_get_bank(mcam, index);
int actbank = bank;
@@ -359,6 +359,41 @@ static void npc_get_keyword(struct mcam_entry *entry, int idx,
*cam0 = ~*cam1 & kw_mask;
}
+static void npc_get_default_entry_action(struct rvu *rvu, struct npc_mcam *mcam,
+ int blkaddr, int index,
+ struct mcam_entry *entry)
+{
+ u16 owner, target_func;
+ struct rvu_pfvf *pfvf;
+ int bank, nixlf;
+ u64 rx_action;
+
+ owner = mcam->entry2pfvf_map[index];
+ target_func = (entry->action >> 4) & 0xffff;
+ /* return incase target is PF or LBK or rule owner is not PF */
+ if (is_afvf(target_func) || (owner & RVU_PFVF_FUNC_MASK) ||
+ !(target_func & RVU_PFVF_FUNC_MASK))
+ return;
+
+ pfvf = rvu_get_pfvf(rvu, target_func);
+ mcam->entry2target_pffunc[index] = target_func;
+ /* return if nixlf is not attached or initialized */
+ if (!is_nixlf_attached(rvu, target_func) || !pfvf->def_ucast_rule)
+ return;
+
+ /* get VF ucast entry rule */
+ nix_get_nixlf(rvu, target_func, &nixlf, NULL);
+ index = npc_get_nixlf_mcam_index(mcam, target_func,
+ nixlf, NIXLF_UCAST_ENTRY);
+ bank = npc_get_bank(mcam, index);
+ index &= (mcam->banksize - 1);
+
+ rx_action = rvu_read64(rvu, blkaddr,
+ NPC_AF_MCAMEX_BANKX_ACTION(index, bank));
+ if (rx_action)
+ entry->action = rx_action;
+}
+
static void npc_config_mcam_entry(struct rvu *rvu, struct npc_mcam *mcam,
int blkaddr, int index, u8 intf,
struct mcam_entry *entry, bool enable)
@@ -406,6 +441,11 @@ static void npc_config_mcam_entry(struct rvu *rvu, struct npc_mcam *mcam,
NPC_AF_MCAMEX_BANKX_CAMX_W1(index, bank, 0), cam0);
}
+ /* copy VF default entry action to the VF mcam entry */
+ if (intf == NIX_INTF_RX && actindex < mcam->bmap_entries)
+ npc_get_default_entry_action(rvu, mcam, blkaddr, actindex,
+ entry);
+
/* Set 'action' */
rvu_write64(rvu, blkaddr,
NPC_AF_MCAMEX_BANKX_ACTION(index, actbank), entry->action);
@@ -473,11 +513,12 @@ void rvu_npc_install_ucast_entry(struct rvu *rvu, u16 pcifunc,
int nixlf, u64 chan, u8 *mac_addr)
{
struct rvu_pfvf *pfvf = rvu_get_pfvf(rvu, pcifunc);
+ u8 mac_mask[] = { [0 ... ETH_ALEN] = 0xFF };
+ struct npc_install_flow_req req = { 0 };
+ struct npc_install_flow_rsp rsp = { 0 };
struct npc_mcam *mcam = &rvu->hw->mcam;
- struct mcam_entry entry = { {0} };
struct nix_rx_action action;
- int blkaddr, index, kwi;
- u64 mac = 0;
+ int blkaddr, index;
/* AF's VFs work in promiscuous mode */
if (is_afvf(pcifunc))
@@ -487,20 +528,9 @@ void rvu_npc_install_ucast_entry(struct rvu *rvu, u16 pcifunc,
if (blkaddr < 0)
return;
- for (index = ETH_ALEN - 1; index >= 0; index--)
- mac |= ((u64)*mac_addr++) << (8 * index);
-
index = npc_get_nixlf_mcam_index(mcam, pcifunc,
nixlf, NIXLF_UCAST_ENTRY);
- /* Match ingress channel and DMAC */
- entry.kw[0] = chan;
- entry.kw_mask[0] = 0xFFFULL;
-
- kwi = NPC_PARSE_RESULT_DMAC_OFFSET / sizeof(u64);
- entry.kw[kwi] = mac;
- entry.kw_mask[kwi] = BIT_ULL(48) - 1;
-
/* Don't change the action if entry is already enabled
* Otherwise RSS action may get overwritten.
*/
@@ -513,20 +543,20 @@ void rvu_npc_install_ucast_entry(struct rvu *rvu, u16 pcifunc,
action.pf_func = pcifunc;
}
- entry.action = *(u64 *)&action;
- npc_config_mcam_entry(rvu, mcam, blkaddr, index,
- pfvf->nix_rx_intf, &entry, true);
-
- /* add VLAN matching, setup action and save entry back for later */
- entry.kw[0] |= (NPC_LT_LB_STAG_QINQ | NPC_LT_LB_CTAG) << 20;
- entry.kw_mask[0] |= (NPC_LT_LB_STAG_QINQ & NPC_LT_LB_CTAG) << 20;
+ req.default_rule = 1;
+ ether_addr_copy(req.packet.dmac, mac_addr);
+ ether_addr_copy(req.mask.dmac, mac_mask);
+ req.features = BIT_ULL(NPC_DMAC);
+ req.channel = chan;
+ req.intf = pfvf->nix_rx_intf;
+ req.op = action.op;
+ req.hdr.pcifunc = 0; /* AF is requester */
+ req.vf = action.pf_func;
+ req.index = action.index;
+ req.match_id = action.match_id;
+ req.flow_key_alg = action.flow_key_alg;
- entry.vtag_action = VTAG0_VALID_BIT |
- FIELD_PREP(VTAG0_TYPE_MASK, 0) |
- FIELD_PREP(VTAG0_LID_MASK, NPC_LID_LA) |
- FIELD_PREP(VTAG0_RELPTR_MASK, 12);
-
- memcpy(&pfvf->entry, &entry, sizeof(entry));
+ rvu_mbox_handler_npc_install_flow(rvu, &req, &rsp);
}
void rvu_npc_install_promisc_entry(struct rvu *rvu, u16 pcifunc,
@@ -682,12 +712,47 @@ void rvu_npc_enable_bcast_entry(struct rvu *rvu, u16 pcifunc, bool enable)
npc_enable_mcam_entry(rvu, mcam, blkaddr, index, enable);
}
+static void npc_update_vf_flow_entry(struct rvu *rvu, struct npc_mcam *mcam,
+ int blkaddr, u16 pcifunc, u64 rx_action)
+{
+ int actindex, index, bank;
+ bool enable;
+
+ if (!(pcifunc & RVU_PFVF_FUNC_MASK))
+ return;
+
+ mutex_lock(&mcam->lock);
+ for (index = 0; index < mcam->bmap_entries; index++) {
+ if (mcam->entry2target_pffunc[index] == pcifunc) {
+ bank = npc_get_bank(mcam, index);
+ actindex = index;
+ index &= (mcam->banksize - 1);
+
+ /* read vf flow entry enable status */
+ enable = is_mcam_entry_enabled(rvu, mcam, blkaddr,
+ actindex);
+ /* disable before mcam entry update */
+ npc_enable_mcam_entry(rvu, mcam, blkaddr, actindex,
+ false);
+ /* update 'action' */
+ rvu_write64(rvu, blkaddr,
+ NPC_AF_MCAMEX_BANKX_ACTION(index, bank),
+ rx_action);
+ if (enable)
+ npc_enable_mcam_entry(rvu, mcam, blkaddr,
+ actindex, true);
+ }
+ }
+ mutex_unlock(&mcam->lock);
+}
+
void rvu_npc_update_flowkey_alg_idx(struct rvu *rvu, u16 pcifunc, int nixlf,
int group, int alg_idx, int mcam_index)
{
struct npc_mcam *mcam = &rvu->hw->mcam;
struct nix_rx_action action;
int blkaddr, index, bank;
+ struct rvu_pfvf *pfvf;
blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NPC, 0);
if (blkaddr < 0)
@@ -724,6 +789,16 @@ void rvu_npc_update_flowkey_alg_idx(struct rvu *rvu, u16 pcifunc, int nixlf,
rvu_write64(rvu, blkaddr,
NPC_AF_MCAMEX_BANKX_ACTION(index, bank), *(u64 *)&action);
+ /* update the VF flow rule action with the VF default entry action */
+ if (mcam_index < 0)
+ npc_update_vf_flow_entry(rvu, mcam, blkaddr, pcifunc,
+ *(u64 *)&action);
+
+ /* update the action change in default rule */
+ pfvf = rvu_get_pfvf(rvu, pcifunc);
+ if (pfvf->def_ucast_rule)
+ pfvf->def_ucast_rule->rx_action = action;
+
index = npc_get_nixlf_mcam_index(mcam, pcifunc,
nixlf, NIXLF_PROMISC_ENTRY);
@@ -738,8 +813,6 @@ void rvu_npc_update_flowkey_alg_idx(struct rvu *rvu, u16 pcifunc, int nixlf,
NPC_AF_MCAMEX_BANKX_ACTION(index, bank),
*(u64 *)&action);
}
-
- rvu_npc_update_rxvlan(rvu, pcifunc, nixlf);
}
static void npc_enadis_default_entries(struct rvu *rvu, u16 pcifunc,
@@ -791,8 +864,6 @@ static void npc_enadis_default_entries(struct rvu *rvu, u16 pcifunc,
rvu_npc_enable_promisc_entry(rvu, pcifunc, nixlf);
else
rvu_npc_disable_promisc_entry(rvu, pcifunc, nixlf);
-
- rvu_npc_update_rxvlan(rvu, pcifunc, nixlf);
}
void rvu_npc_disable_default_entries(struct rvu *rvu, u16 pcifunc, int nixlf)
@@ -807,7 +878,9 @@ void rvu_npc_enable_default_entries(struct rvu *rvu, u16 pcifunc, int nixlf)
void rvu_npc_disable_mcam_entries(struct rvu *rvu, u16 pcifunc, int nixlf)
{
+ struct rvu_pfvf *pfvf = rvu_get_pfvf(rvu, pcifunc);
struct npc_mcam *mcam = &rvu->hw->mcam;
+ struct rvu_npc_mcam_rule *rule;
int blkaddr;
blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NPC, 0);
@@ -816,12 +889,52 @@ void rvu_npc_disable_mcam_entries(struct rvu *rvu, u16 pcifunc, int nixlf)
mutex_lock(&mcam->lock);
- /* Disable and free all MCAM entries mapped to this 'pcifunc' */
+ /* Disable MCAM entries directing traffic to this 'pcifunc' */
+ list_for_each_entry(rule, &mcam->mcam_rules, list) {
+ if (is_npc_intf_rx(rule->intf) &&
+ rule->rx_action.pf_func == pcifunc) {
+ npc_enable_mcam_entry(rvu, mcam, blkaddr,
+ rule->entry, false);
+ rule->enable = false;
+ /* Indicate that default rule is disabled */
+ if (rule->default_rule)
+ pfvf->def_ucast_rule = NULL;
+ }
+ }
+
+ mutex_unlock(&mcam->lock);
+
+ npc_mcam_disable_flows(rvu, pcifunc);
+
+ rvu_npc_disable_default_entries(rvu, pcifunc, nixlf);
+}
+
+void rvu_npc_free_mcam_entries(struct rvu *rvu, u16 pcifunc, int nixlf)
+{
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ struct rvu_npc_mcam_rule *rule, *tmp;
+ int blkaddr;
+
+ blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NPC, 0);
+ if (blkaddr < 0)
+ return;
+
+ mutex_lock(&mcam->lock);
+
+ /* Free all MCAM entries owned by this 'pcifunc' */
npc_mcam_free_all_entries(rvu, mcam, blkaddr, pcifunc);
- /* Free all MCAM counters mapped to this 'pcifunc' */
+ /* Free all MCAM counters owned by this 'pcifunc' */
npc_mcam_free_all_counters(rvu, mcam, pcifunc);
+ /* Delete MCAM entries owned by this 'pcifunc' */
+ list_for_each_entry_safe(rule, tmp, &mcam->mcam_rules, list) {
+ if (rule->owner == pcifunc && !rule->default_rule) {
+ list_del(&rule->list);
+ kfree(rule);
+ }
+ }
+
mutex_unlock(&mcam->lock);
rvu_npc_disable_default_entries(rvu, pcifunc, nixlf);
@@ -1231,6 +1344,12 @@ static int npc_mcam_rsrcs_init(struct rvu *rvu, int blkaddr)
if (!mcam->cntr_refcnt)
goto free_mem;
+ /* Alloc memory for saving target device of mcam rule */
+ mcam->entry2target_pffunc = devm_kcalloc(rvu->dev, mcam->total_entries,
+ sizeof(u16), GFP_KERNEL);
+ if (!mcam->entry2target_pffunc)
+ goto free_mem;
+
mutex_init(&mcam->lock);
return 0;
@@ -1580,6 +1699,7 @@ static void npc_mcam_free_all_entries(struct rvu *rvu, struct npc_mcam *mcam,
npc_unmap_mcam_entry_and_cntr(rvu, mcam,
blkaddr, index,
cntr);
+ mcam->entry2target_pffunc[index] = 0x0;
}
}
}
@@ -1966,6 +2086,7 @@ int rvu_mbox_handler_npc_mcam_free_entry(struct rvu *rvu,
goto exit;
mcam->entry2pfvf_map[req->entry] = 0;
+ mcam->entry2target_pffunc[req->entry] = 0x0;
npc_mcam_clear_bit(mcam, req->entry);
npc_enable_mcam_entry(rvu, mcam, blkaddr, req->entry, false);
@@ -2521,3 +2642,27 @@ int rvu_npc_update_rxvlan(struct rvu *rvu, u16 pcifunc, int nixlf)
return 0;
}
+
+bool rvu_npc_write_default_rule(struct rvu *rvu, int blkaddr, int nixlf,
+ u16 pcifunc, u8 intf, struct mcam_entry *entry,
+ int *index)
+{
+ struct rvu_pfvf *pfvf = rvu_get_pfvf(rvu, pcifunc);
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ bool enable;
+ u8 nix_intf;
+
+ if (is_npc_intf_tx(intf))
+ nix_intf = pfvf->nix_tx_intf;
+ else
+ nix_intf = pfvf->nix_rx_intf;
+
+ *index = npc_get_nixlf_mcam_index(mcam, pcifunc,
+ nixlf, NIXLF_UCAST_ENTRY);
+ /* dont force enable unicast entry */
+ enable = is_mcam_entry_enabled(rvu, mcam, blkaddr, *index);
+ npc_config_mcam_entry(rvu, mcam, blkaddr, *index, nix_intf,
+ entry, enable);
+
+ return enable;
+}
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c
index 4dae89776422..94b52dfb6ac6 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c
@@ -558,5 +558,726 @@ static int npc_scan_verify_kex(struct rvu *rvu, int blkaddr)
int npc_flow_steering_init(struct rvu *rvu, int blkaddr)
{
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+
+ INIT_LIST_HEAD(&mcam->mcam_rules);
+
return npc_scan_verify_kex(rvu, blkaddr);
}
+
+static int npc_check_unsupported_flows(struct rvu *rvu, u64 features, u8 intf)
+{
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ u64 *mcam_features = &mcam->rx_features;
+ u64 unsupported;
+ u8 bit;
+
+ if (is_npc_intf_tx(intf))
+ mcam_features = &mcam->tx_features;
+
+ unsupported = (*mcam_features ^ features) & ~(*mcam_features);
+ if (unsupported) {
+ dev_info(rvu->dev, "Unsupported flow(s):\n");
+ for_each_set_bit(bit, (unsigned long *)&unsupported, 64)
+ dev_info(rvu->dev, "%s ", npc_get_field_name(bit));
+ return -EOPNOTSUPP;
+ }
+
+ return 0;
+}
+
+/* npc_update_entry - Based on the masks generated during
+ * the key scanning, updates the given entry with value and
+ * masks for the field of interest. Maximum 16 bytes of a packet
+ * header can be extracted by HW hence lo and hi are sufficient.
+ * When field bytes are less than or equal to 8 then hi should be
+ * 0 for value and mask.
+ *
+ * If exact match of value is required then mask should be all 1's.
+ * If any bits in mask are 0 then corresponding bits in value are
+ * dont care.
+ */
+static void npc_update_entry(struct rvu *rvu, enum key_fields type,
+ struct mcam_entry *entry, u64 val_lo,
+ u64 val_hi, u64 mask_lo, u64 mask_hi, u8 intf)
+{
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ struct mcam_entry dummy = { {0} };
+ struct npc_key_field *field;
+ u64 kw1, kw2, kw3;
+ u8 shift;
+ int i;
+
+ field = &mcam->rx_key_fields[type];
+ if (is_npc_intf_tx(intf))
+ field = &mcam->tx_key_fields[type];
+
+ if (!field->nr_kws)
+ return;
+
+ for (i = 0; i < NPC_MAX_KWS_IN_KEY; i++) {
+ if (!field->kw_mask[i])
+ continue;
+ /* place key value in kw[x] */
+ shift = __ffs64(field->kw_mask[i]);
+ /* update entry value */
+ kw1 = (val_lo << shift) & field->kw_mask[i];
+ dummy.kw[i] = kw1;
+ /* update entry mask */
+ kw1 = (mask_lo << shift) & field->kw_mask[i];
+ dummy.kw_mask[i] = kw1;
+
+ if (field->nr_kws == 1)
+ break;
+ /* place remaining bits of key value in kw[x + 1] */
+ if (field->nr_kws == 2) {
+ /* update entry value */
+ kw2 = shift ? val_lo >> (64 - shift) : 0;
+ kw2 |= (val_hi << shift);
+ kw2 &= field->kw_mask[i + 1];
+ dummy.kw[i + 1] = kw2;
+ /* update entry mask */
+ kw2 = shift ? mask_lo >> (64 - shift) : 0;
+ kw2 |= (mask_hi << shift);
+ kw2 &= field->kw_mask[i + 1];
+ dummy.kw_mask[i + 1] = kw2;
+ break;
+ }
+ /* place remaining bits of key value in kw[x + 1], kw[x + 2] */
+ if (field->nr_kws == 3) {
+ /* update entry value */
+ kw2 = shift ? val_lo >> (64 - shift) : 0;
+ kw2 |= (val_hi << shift);
+ kw2 &= field->kw_mask[i + 1];
+ kw3 = shift ? val_hi >> (64 - shift) : 0;
+ kw3 &= field->kw_mask[i + 2];
+ dummy.kw[i + 1] = kw2;
+ dummy.kw[i + 2] = kw3;
+ /* update entry mask */
+ kw2 = shift ? mask_lo >> (64 - shift) : 0;
+ kw2 |= (mask_hi << shift);
+ kw2 &= field->kw_mask[i + 1];
+ kw3 = shift ? mask_hi >> (64 - shift) : 0;
+ kw3 &= field->kw_mask[i + 2];
+ dummy.kw_mask[i + 1] = kw2;
+ dummy.kw_mask[i + 2] = kw3;
+ break;
+ }
+ }
+ /* dummy is ready with values and masks for given key
+ * field now clear and update input entry with those
+ */
+ for (i = 0; i < NPC_MAX_KWS_IN_KEY; i++) {
+ if (!field->kw_mask[i])
+ continue;
+ entry->kw[i] &= ~field->kw_mask[i];
+ entry->kw_mask[i] &= ~field->kw_mask[i];
+
+ entry->kw[i] |= dummy.kw[i];
+ entry->kw_mask[i] |= dummy.kw_mask[i];
+ }
+}
+
+#define IPV6_WORDS 4
+
+static void npc_update_ipv6_flow(struct rvu *rvu, struct mcam_entry *entry,
+ u64 features, struct flow_msg *pkt,
+ struct flow_msg *mask,
+ struct rvu_npc_mcam_rule *output, u8 intf)
+{
+ u32 src_ip[IPV6_WORDS], src_ip_mask[IPV6_WORDS];
+ u32 dst_ip[IPV6_WORDS], dst_ip_mask[IPV6_WORDS];
+ struct flow_msg *opkt = &output->packet;
+ struct flow_msg *omask = &output->mask;
+ u64 mask_lo, mask_hi;
+ u64 val_lo, val_hi;
+
+ /* For an ipv6 address fe80::2c68:63ff:fe5e:2d0a the packet
+ * values to be programmed in MCAM should as below:
+ * val_high: 0xfe80000000000000
+ * val_low: 0x2c6863fffe5e2d0a
+ */
+ if (features & BIT_ULL(NPC_SIP_IPV6)) {
+ be32_to_cpu_array(src_ip_mask, mask->ip6src, IPV6_WORDS);
+ be32_to_cpu_array(src_ip, pkt->ip6src, IPV6_WORDS);
+
+ mask_hi = (u64)src_ip_mask[0] << 32 | src_ip_mask[1];
+ mask_lo = (u64)src_ip_mask[2] << 32 | src_ip_mask[3];
+ val_hi = (u64)src_ip[0] << 32 | src_ip[1];
+ val_lo = (u64)src_ip[2] << 32 | src_ip[3];
+
+ npc_update_entry(rvu, NPC_SIP_IPV6, entry, val_lo, val_hi,
+ mask_lo, mask_hi, intf);
+ memcpy(opkt->ip6src, pkt->ip6src, sizeof(opkt->ip6src));
+ memcpy(omask->ip6src, mask->ip6src, sizeof(omask->ip6src));
+ }
+ if (features & BIT_ULL(NPC_DIP_IPV6)) {
+ be32_to_cpu_array(dst_ip_mask, mask->ip6dst, IPV6_WORDS);
+ be32_to_cpu_array(dst_ip, pkt->ip6dst, IPV6_WORDS);
+
+ mask_hi = (u64)dst_ip_mask[0] << 32 | dst_ip_mask[1];
+ mask_lo = (u64)dst_ip_mask[2] << 32 | dst_ip_mask[3];
+ val_hi = (u64)dst_ip[0] << 32 | dst_ip[1];
+ val_lo = (u64)dst_ip[2] << 32 | dst_ip[3];
+
+ npc_update_entry(rvu, NPC_DIP_IPV6, entry, val_lo, val_hi,
+ mask_lo, mask_hi, intf);
+ memcpy(opkt->ip6dst, pkt->ip6dst, sizeof(opkt->ip6dst));
+ memcpy(omask->ip6dst, mask->ip6dst, sizeof(omask->ip6dst));
+ }
+}
+
+static void npc_update_flow(struct rvu *rvu, struct mcam_entry *entry,
+ u64 features, struct flow_msg *pkt,
+ struct flow_msg *mask,
+ struct rvu_npc_mcam_rule *output, u8 intf)
+{
+ u64 dmac_mask = ether_addr_to_u64(mask->dmac);
+ u64 smac_mask = ether_addr_to_u64(mask->smac);
+ u64 dmac_val = ether_addr_to_u64(pkt->dmac);
+ u64 smac_val = ether_addr_to_u64(pkt->smac);
+ struct flow_msg *opkt = &output->packet;
+ struct flow_msg *omask = &output->mask;
+
+ if (!features)
+ return;
+
+#define NPC_WRITE_FLOW(field, member, val_lo, val_hi, mask_lo, mask_hi) \
+do { \
+ if (features & BIT_ULL((field))) { \
+ npc_update_entry(rvu, (field), entry, (val_lo), (val_hi), \
+ (mask_lo), (mask_hi), intf); \
+ memcpy(&opkt->member, &pkt->member, sizeof(pkt->member)); \
+ memcpy(&omask->member, &mask->member, sizeof(mask->member)); \
+ } \
+} while (0)
+
+ /* For tcp/udp/sctp LTYPE should be present in entry */
+ if (features & (BIT_ULL(NPC_SPORT_TCP) | BIT_ULL(NPC_DPORT_TCP)))
+ npc_update_entry(rvu, NPC_LD, entry, NPC_LT_LD_TCP,
+ 0, ~0ULL, 0, intf);
+ if (features & (BIT_ULL(NPC_SPORT_UDP) | BIT_ULL(NPC_DPORT_UDP)))
+ npc_update_entry(rvu, NPC_LD, entry, NPC_LT_LD_UDP,
+ 0, ~0ULL, 0, intf);
+ if (features & (BIT_ULL(NPC_SPORT_SCTP) | BIT_ULL(NPC_DPORT_SCTP)))
+ npc_update_entry(rvu, NPC_LD, entry, NPC_LT_LD_SCTP,
+ 0, ~0ULL, 0, intf);
+
+ if (features & BIT_ULL(NPC_OUTER_VID))
+ npc_update_entry(rvu, NPC_LB, entry,
+ NPC_LT_LB_STAG_QINQ | NPC_LT_LB_CTAG, 0,
+ NPC_LT_LB_STAG_QINQ & NPC_LT_LB_CTAG, 0, intf);
+
+ NPC_WRITE_FLOW(NPC_DMAC, dmac, dmac_val, 0, dmac_mask, 0);
+ NPC_WRITE_FLOW(NPC_SMAC, smac, smac_val, 0, smac_mask, 0);
+ NPC_WRITE_FLOW(NPC_ETYPE, etype, ntohs(pkt->etype), 0,
+ ntohs(mask->etype), 0);
+ NPC_WRITE_FLOW(NPC_SIP_IPV4, ip4src, ntohl(pkt->ip4src), 0,
+ ntohl(mask->ip4src), 0);
+ NPC_WRITE_FLOW(NPC_DIP_IPV4, ip4dst, ntohl(pkt->ip4dst), 0,
+ ntohl(mask->ip4dst), 0);
+ NPC_WRITE_FLOW(NPC_SPORT_TCP, sport, ntohs(pkt->sport), 0,
+ ntohs(mask->sport), 0);
+ NPC_WRITE_FLOW(NPC_SPORT_UDP, sport, ntohs(pkt->sport), 0,
+ ntohs(mask->sport), 0);
+ NPC_WRITE_FLOW(NPC_DPORT_TCP, dport, ntohs(pkt->dport), 0,
+ ntohs(mask->dport), 0);
+ NPC_WRITE_FLOW(NPC_DPORT_UDP, dport, ntohs(pkt->dport), 0,
+ ntohs(mask->dport), 0);
+ NPC_WRITE_FLOW(NPC_SPORT_SCTP, sport, ntohs(pkt->sport), 0,
+ ntohs(mask->sport), 0);
+ NPC_WRITE_FLOW(NPC_DPORT_SCTP, dport, ntohs(pkt->dport), 0,
+ ntohs(mask->dport), 0);
+
+ NPC_WRITE_FLOW(NPC_OUTER_VID, vlan_tci, ntohs(pkt->vlan_tci), 0,
+ ntohs(mask->vlan_tci), 0);
+
+ npc_update_ipv6_flow(rvu, entry, features, pkt, mask, output, intf);
+}
+
+static struct rvu_npc_mcam_rule *rvu_mcam_find_rule(struct npc_mcam *mcam,
+ u16 entry)
+{
+ struct rvu_npc_mcam_rule *iter;
+
+ mutex_lock(&mcam->lock);
+ list_for_each_entry(iter, &mcam->mcam_rules, list) {
+ if (iter->entry == entry) {
+ mutex_unlock(&mcam->lock);
+ return iter;
+ }
+ }
+ mutex_unlock(&mcam->lock);
+
+ return NULL;
+}
+
+static void rvu_mcam_add_rule(struct npc_mcam *mcam,
+ struct rvu_npc_mcam_rule *rule)
+{
+ struct list_head *head = &mcam->mcam_rules;
+ struct rvu_npc_mcam_rule *iter;
+
+ mutex_lock(&mcam->lock);
+ list_for_each_entry(iter, &mcam->mcam_rules, list) {
+ if (iter->entry > rule->entry)
+ break;
+ head = &iter->list;
+ }
+
+ list_add(&rule->list, head);
+ mutex_unlock(&mcam->lock);
+}
+
+static void rvu_mcam_remove_counter_from_rule(struct rvu *rvu, u16 pcifunc,
+ struct rvu_npc_mcam_rule *rule)
+{
+ struct npc_mcam_oper_counter_req free_req = { 0 };
+ struct msg_rsp free_rsp;
+
+ if (!rule->has_cntr)
+ return;
+
+ free_req.hdr.pcifunc = pcifunc;
+ free_req.cntr = rule->cntr;
+
+ rvu_mbox_handler_npc_mcam_free_counter(rvu, &free_req, &free_rsp);
+ rule->has_cntr = false;
+}
+
+static void rvu_mcam_add_counter_to_rule(struct rvu *rvu, u16 pcifunc,
+ struct rvu_npc_mcam_rule *rule,
+ struct npc_install_flow_rsp *rsp)
+{
+ struct npc_mcam_alloc_counter_req cntr_req = { 0 };
+ struct npc_mcam_alloc_counter_rsp cntr_rsp = { 0 };
+ int err;
+
+ cntr_req.hdr.pcifunc = pcifunc;
+ cntr_req.contig = true;
+ cntr_req.count = 1;
+
+ /* we try to allocate a counter to track the stats of this
+ * rule. If counter could not be allocated then proceed
+ * without counter because counters are limited than entries.
+ */
+ err = rvu_mbox_handler_npc_mcam_alloc_counter(rvu, &cntr_req,
+ &cntr_rsp);
+ if (!err && cntr_rsp.count) {
+ rule->cntr = cntr_rsp.cntr;
+ rule->has_cntr = true;
+ rsp->counter = rule->cntr;
+ } else {
+ rsp->counter = err;
+ }
+}
+
+static void npc_update_rx_entry(struct rvu *rvu, struct rvu_pfvf *pfvf,
+ struct mcam_entry *entry,
+ struct npc_install_flow_req *req, u16 target)
+{
+ struct nix_rx_action action;
+
+ npc_update_entry(rvu, NPC_CHAN, entry, req->channel, 0,
+ ~0ULL, 0, NIX_INTF_RX);
+
+ *(u64 *)&action = 0x00;
+ action.pf_func = target;
+ action.op = req->op;
+ action.index = req->index;
+ action.match_id = req->match_id;
+ action.flow_key_alg = req->flow_key_alg;
+
+ if (req->op == NIX_RX_ACTION_DEFAULT && pfvf->def_ucast_rule)
+ action = pfvf->def_ucast_rule->rx_action;
+
+ entry->action = *(u64 *)&action;
+
+ /* VTAG0 starts at 0th byte of LID_B.
+ * VTAG1 starts at 4th byte of LID_B.
+ */
+ entry->vtag_action = FIELD_PREP(RX_VTAG0_VALID_BIT, req->vtag0_valid) |
+ FIELD_PREP(RX_VTAG0_TYPE_MASK, req->vtag0_type) |
+ FIELD_PREP(RX_VTAG0_LID_MASK, NPC_LID_LB) |
+ FIELD_PREP(RX_VTAG0_RELPTR_MASK, 0) |
+ FIELD_PREP(RX_VTAG1_VALID_BIT, req->vtag1_valid) |
+ FIELD_PREP(RX_VTAG1_TYPE_MASK, req->vtag1_type) |
+ FIELD_PREP(RX_VTAG1_LID_MASK, NPC_LID_LB) |
+ FIELD_PREP(RX_VTAG1_RELPTR_MASK, 4);
+}
+
+static void npc_update_tx_entry(struct rvu *rvu, struct rvu_pfvf *pfvf,
+ struct mcam_entry *entry,
+ struct npc_install_flow_req *req, u16 target)
+{
+ struct nix_tx_action action;
+
+ npc_update_entry(rvu, NPC_PF_FUNC, entry, (__force u16)htons(target),
+ 0, ~0ULL, 0, NIX_INTF_TX);
+
+ *(u64 *)&action = 0x00;
+ action.op = req->op;
+ action.index = req->index;
+ action.match_id = req->match_id;
+
+ entry->action = *(u64 *)&action;
+
+ /* VTAG0 starts at 0th byte of LID_B.
+ * VTAG1 starts at 4th byte of LID_B.
+ */
+ entry->vtag_action = FIELD_PREP(TX_VTAG0_DEF_MASK, req->vtag0_def) |
+ FIELD_PREP(TX_VTAG0_OP_MASK, req->vtag0_op) |
+ FIELD_PREP(TX_VTAG0_LID_MASK, NPC_LID_LA) |
+ FIELD_PREP(TX_VTAG0_RELPTR_MASK, 20) |
+ FIELD_PREP(TX_VTAG1_DEF_MASK, req->vtag1_def) |
+ FIELD_PREP(TX_VTAG1_OP_MASK, req->vtag1_op) |
+ FIELD_PREP(TX_VTAG1_LID_MASK, NPC_LID_LA) |
+ FIELD_PREP(TX_VTAG1_RELPTR_MASK, 24);
+}
+
+static int npc_install_flow(struct rvu *rvu, int blkaddr, u16 target,
+ int nixlf, struct rvu_pfvf *pfvf,
+ struct npc_install_flow_req *req,
+ struct npc_install_flow_rsp *rsp, bool enable)
+{
+ struct rvu_npc_mcam_rule *def_ucast_rule = pfvf->def_ucast_rule;
+ u64 features, installed_features, missing_features = 0;
+ struct npc_mcam_write_entry_req write_req = { 0 };
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ struct rvu_npc_mcam_rule dummy = { 0 };
+ struct rvu_npc_mcam_rule *rule;
+ bool new = false, msg_from_vf;
+ u16 owner = req->hdr.pcifunc;
+ struct msg_rsp write_rsp;
+ struct mcam_entry *entry;
+ int entry_index, err;
+
+ msg_from_vf = !!(owner & RVU_PFVF_FUNC_MASK);
+
+ installed_features = req->features;
+ features = req->features;
+ entry = &write_req.entry_data;
+ entry_index = req->entry;
+
+ npc_update_flow(rvu, entry, features, &req->packet, &req->mask, &dummy,
+ req->intf);
+
+ if (is_npc_intf_rx(req->intf))
+ npc_update_rx_entry(rvu, pfvf, entry, req, target);
+ else
+ npc_update_tx_entry(rvu, pfvf, entry, req, target);
+
+ /* Default unicast rules do not exist for TX */
+ if (is_npc_intf_tx(req->intf))
+ goto find_rule;
+
+ if (def_ucast_rule)
+ missing_features = (def_ucast_rule->features ^ features) &
+ def_ucast_rule->features;
+
+ if (req->default_rule && req->append) {
+ /* add to default rule */
+ if (missing_features)
+ npc_update_flow(rvu, entry, missing_features,
+ &def_ucast_rule->packet,
+ &def_ucast_rule->mask,
+ &dummy, req->intf);
+ enable = rvu_npc_write_default_rule(rvu, blkaddr,
+ nixlf, target,
+ pfvf->nix_rx_intf, entry,
+ &entry_index);
+ installed_features = req->features | missing_features;
+ } else if (req->default_rule && !req->append) {
+ /* overwrite default rule */
+ enable = rvu_npc_write_default_rule(rvu, blkaddr,
+ nixlf, target,
+ pfvf->nix_rx_intf, entry,
+ &entry_index);
+ } else if (msg_from_vf) {
+ /* normal rule - include default rule also to it for VF */
+ npc_update_flow(rvu, entry, missing_features,
+ &def_ucast_rule->packet, &def_ucast_rule->mask,
+ &dummy, req->intf);
+ installed_features = req->features | missing_features;
+ }
+
+find_rule:
+ rule = rvu_mcam_find_rule(mcam, entry_index);
+ if (!rule) {
+ rule = kzalloc(sizeof(*rule), GFP_KERNEL);
+ if (!rule)
+ return -ENOMEM;
+ new = true;
+ }
+ /* no counter for default rule */
+ if (req->default_rule)
+ goto update_rule;
+
+ /* allocate new counter if rule has no counter */
+ if (req->set_cntr && !rule->has_cntr)
+ rvu_mcam_add_counter_to_rule(rvu, owner, rule, rsp);
+
+ /* if user wants to delete an existing counter for a rule then
+ * free the counter
+ */
+ if (!req->set_cntr && rule->has_cntr)
+ rvu_mcam_remove_counter_from_rule(rvu, owner, rule);
+
+ write_req.hdr.pcifunc = owner;
+ write_req.entry = req->entry;
+ write_req.intf = req->intf;
+ write_req.enable_entry = (u8)enable;
+ /* if counter is available then clear and use it */
+ if (req->set_cntr && rule->has_cntr) {
+ rvu_write64(rvu, blkaddr, NPC_AF_MATCH_STATX(rule->cntr), 0x00);
+ write_req.set_cntr = 1;
+ write_req.cntr = rule->cntr;
+ }
+
+ err = rvu_mbox_handler_npc_mcam_write_entry(rvu, &write_req,
+ &write_rsp);
+ if (err) {
+ rvu_mcam_remove_counter_from_rule(rvu, owner, rule);
+ if (new)
+ kfree(rule);
+ return err;
+ }
+update_rule:
+ memcpy(&rule->packet, &dummy.packet, sizeof(rule->packet));
+ memcpy(&rule->mask, &dummy.mask, sizeof(rule->mask));
+ rule->entry = entry_index;
+ memcpy(&rule->rx_action, &entry->action, sizeof(struct nix_rx_action));
+ if (is_npc_intf_tx(req->intf))
+ memcpy(&rule->tx_action, &entry->action,
+ sizeof(struct nix_tx_action));
+ rule->vtag_action = entry->vtag_action;
+ rule->features = installed_features;
+ rule->default_rule = req->default_rule;
+ rule->owner = owner;
+ rule->enable = enable;
+ if (is_npc_intf_tx(req->intf))
+ rule->intf = pfvf->nix_tx_intf;
+ else
+ rule->intf = pfvf->nix_rx_intf;
+
+ if (new)
+ rvu_mcam_add_rule(mcam, rule);
+ if (req->default_rule)
+ pfvf->def_ucast_rule = rule;
+
+ return 0;
+}
+
+int rvu_mbox_handler_npc_install_flow(struct rvu *rvu,
+ struct npc_install_flow_req *req,
+ struct npc_install_flow_rsp *rsp)
+{
+ bool from_vf = !!(req->hdr.pcifunc & RVU_PFVF_FUNC_MASK);
+ int blkaddr, nixlf, err;
+ struct rvu_pfvf *pfvf;
+ bool enable = true;
+ u16 target;
+
+ blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NPC, 0);
+ if (blkaddr < 0) {
+ dev_err(rvu->dev, "%s: NPC block not implemented\n", __func__);
+ return -ENODEV;
+ }
+
+ if (!is_npc_interface_valid(rvu, req->intf))
+ return -EINVAL;
+
+ if (from_vf && req->default_rule)
+ return NPC_MCAM_PERM_DENIED;
+
+ /* Each PF/VF info is maintained in struct rvu_pfvf.
+ * rvu_pfvf for the target PF/VF needs to be retrieved
+ * hence modify pcifunc accordingly.
+ */
+
+ /* AF installing for a PF/VF */
+ if (!req->hdr.pcifunc)
+ target = req->vf;
+ /* PF installing for its VF */
+ else if (!from_vf && req->vf)
+ target = (req->hdr.pcifunc & ~RVU_PFVF_FUNC_MASK) | req->vf;
+ /* msg received from PF/VF */
+ else
+ target = req->hdr.pcifunc;
+
+ if (npc_check_unsupported_flows(rvu, req->features, req->intf))
+ return -EOPNOTSUPP;
+
+ if (npc_mcam_verify_channel(rvu, target, req->intf, req->channel))
+ return -EINVAL;
+
+ pfvf = rvu_get_pfvf(rvu, target);
+
+ /* update req destination mac addr */
+ if ((req->features & BIT_ULL(NPC_DMAC)) && is_npc_intf_rx(req->intf) &&
+ is_zero_ether_addr(req->packet.dmac)) {
+ ether_addr_copy(req->packet.dmac, pfvf->mac_addr);
+ eth_broadcast_addr((u8 *)&req->mask.dmac);
+ }
+
+ err = nix_get_nixlf(rvu, target, &nixlf, NULL);
+
+ /* If interface is uninitialized then do not enable entry */
+ if (err || (!req->default_rule && !pfvf->def_ucast_rule))
+ enable = false;
+
+ /* Packets reaching NPC in Tx path implies that a
+ * NIXLF is properly setup and transmitting.
+ * Hence rules can be enabled for Tx.
+ */
+ if (is_npc_intf_tx(req->intf))
+ enable = true;
+
+ /* Do not allow requests from uninitialized VFs */
+ if (from_vf && !enable)
+ return -EINVAL;
+
+ /* If message is from VF then its flow should not overlap with
+ * reserved unicast flow.
+ */
+ if (from_vf && pfvf->def_ucast_rule && is_npc_intf_rx(req->intf) &&
+ pfvf->def_ucast_rule->features & req->features)
+ return -EINVAL;
+
+ return npc_install_flow(rvu, blkaddr, target, nixlf, pfvf, req, rsp,
+ enable);
+}
+
+static int npc_delete_flow(struct rvu *rvu, struct rvu_npc_mcam_rule *rule,
+ u16 pcifunc)
+{
+ struct npc_mcam_ena_dis_entry_req dis_req = { 0 };
+ struct msg_rsp dis_rsp;
+
+ if (rule->default_rule)
+ return 0;
+
+ if (rule->has_cntr)
+ rvu_mcam_remove_counter_from_rule(rvu, pcifunc, rule);
+
+ dis_req.hdr.pcifunc = pcifunc;
+ dis_req.entry = rule->entry;
+
+ list_del(&rule->list);
+ kfree(rule);
+
+ return rvu_mbox_handler_npc_mcam_dis_entry(rvu, &dis_req, &dis_rsp);
+}
+
+int rvu_mbox_handler_npc_delete_flow(struct rvu *rvu,
+ struct npc_delete_flow_req *req,
+ struct msg_rsp *rsp)
+{
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ struct rvu_npc_mcam_rule *iter, *tmp;
+ u16 pcifunc = req->hdr.pcifunc;
+ struct list_head del_list;
+
+ INIT_LIST_HEAD(&del_list);
+
+ mutex_lock(&mcam->lock);
+ list_for_each_entry_safe(iter, tmp, &mcam->mcam_rules, list) {
+ if (iter->owner == pcifunc) {
+ /* All rules */
+ if (req->all) {
+ list_move_tail(&iter->list, &del_list);
+ /* Range of rules */
+ } else if (req->end && iter->entry >= req->start &&
+ iter->entry <= req->end) {
+ list_move_tail(&iter->list, &del_list);
+ /* single rule */
+ } else if (req->entry == iter->entry) {
+ list_move_tail(&iter->list, &del_list);
+ break;
+ }
+ }
+ }
+ mutex_unlock(&mcam->lock);
+
+ list_for_each_entry_safe(iter, tmp, &del_list, list) {
+ /* clear the mcam entry target pcifunc */
+ mcam->entry2target_pffunc[iter->entry] = 0x0;
+ if (npc_delete_flow(rvu, iter, pcifunc))
+ dev_err(rvu->dev, "rule deletion failed for entry:%d",
+ iter->entry);
+ }
+
+ return 0;
+}
+
+void npc_mcam_enable_flows(struct rvu *rvu, u16 target)
+{
+ struct rvu_pfvf *pfvf = rvu_get_pfvf(rvu, target);
+ struct rvu_npc_mcam_rule *def_ucast_rule;
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ struct rvu_npc_mcam_rule *rule;
+ int blkaddr, bank, index;
+ u64 def_action;
+
+ blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NPC, 0);
+ if (blkaddr < 0)
+ return;
+
+ def_ucast_rule = pfvf->def_ucast_rule;
+
+ mutex_lock(&mcam->lock);
+ list_for_each_entry(rule, &mcam->mcam_rules, list) {
+ if (is_npc_intf_rx(rule->intf) &&
+ rule->rx_action.pf_func == target && !rule->enable) {
+ if (rule->default_rule) {
+ npc_enable_mcam_entry(rvu, mcam, blkaddr,
+ rule->entry, true);
+ rule->enable = true;
+ continue;
+ }
+
+ if (rule->rx_action.op == NIX_RX_ACTION_DEFAULT) {
+ if (!def_ucast_rule)
+ continue;
+ /* Use default unicast entry action */
+ rule->rx_action = def_ucast_rule->rx_action;
+ def_action = *(u64 *)&def_ucast_rule->rx_action;
+ bank = npc_get_bank(mcam, rule->entry);
+ rvu_write64(rvu, blkaddr,
+ NPC_AF_MCAMEX_BANKX_ACTION
+ (rule->entry, bank), def_action);
+ }
+
+ npc_enable_mcam_entry(rvu, mcam, blkaddr,
+ rule->entry, true);
+ rule->enable = true;
+ }
+ }
+
+ /* Enable MCAM entries installed by PF with target as VF pcifunc */
+ for (index = 0; index < mcam->bmap_entries; index++) {
+ if (mcam->entry2target_pffunc[index] == target)
+ npc_enable_mcam_entry(rvu, mcam, blkaddr,
+ index, true);
+ }
+ mutex_unlock(&mcam->lock);
+}
+
+void npc_mcam_disable_flows(struct rvu *rvu, u16 target)
+{
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ int blkaddr, index;
+
+ blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NPC, 0);
+ if (blkaddr < 0)
+ return;
+
+ mutex_lock(&mcam->lock);
+ /* Disable MCAM entries installed by PF with target as VF pcifunc */
+ for (index = 0; index < mcam->bmap_entries; index++) {
+ if (mcam->entry2target_pffunc[index] == target)
+ npc_enable_mcam_entry(rvu, mcam, blkaddr,
+ index, false);
+ }
+ mutex_unlock(&mcam->lock);
+}
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index 66f1a212f1f4..672768630557 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -1278,6 +1278,7 @@ static void otx2_free_sq_res(struct otx2_nic *pf)
static int otx2_init_hw_resources(struct otx2_nic *pf)
{
+ struct nix_lf_free_req *free_req;
struct mbox *mbox = &pf->mbox;
struct otx2_hw *hw = &pf->hw;
struct msg_req *req;
@@ -1359,8 +1360,9 @@ static int otx2_init_hw_resources(struct otx2_nic *pf)
otx2_aura_pool_free(pf);
err_free_nix_lf:
mutex_lock(&mbox->lock);
- req = otx2_mbox_alloc_msg_nix_lf_free(mbox);
- if (req) {
+ free_req = otx2_mbox_alloc_msg_nix_lf_free(mbox);
+ if (free_req) {
+ free_req->flags = NIX_LF_DISABLE_FLOWS;
if (otx2_sync_mbox_msg(mbox))
dev_err(pf->dev, "%s failed to free nixlf\n", __func__);
}
@@ -1379,6 +1381,7 @@ static int otx2_init_hw_resources(struct otx2_nic *pf)
static void otx2_free_hw_resources(struct otx2_nic *pf)
{
struct otx2_qset *qset = &pf->qset;
+ struct nix_lf_free_req *free_req;
struct mbox *mbox = &pf->mbox;
struct otx2_cq_queue *cq;
struct msg_req *req;
@@ -1419,8 +1422,9 @@ static void otx2_free_hw_resources(struct otx2_nic *pf)
mutex_lock(&mbox->lock);
/* Reset NIX LF */
- req = otx2_mbox_alloc_msg_nix_lf_free(mbox);
- if (req) {
+ free_req = otx2_mbox_alloc_msg_nix_lf_free(mbox);
+ if (free_req) {
+ free_req->flags = NIX_LF_DISABLE_FLOWS;
if (otx2_sync_mbox_msg(mbox))
dev_err(pf->dev, "%s failed to free nixlf\n", __func__);
}
--
2.16.5
^ permalink raw reply related [flat|nested] 24+ messages in thread* Re: [PATCH v3 net-next 04/13] octeontx2-af: Add mbox messages to install and delete MCAM rules
2020-11-11 7:13 ` [PATCH v3 net-next 04/13] octeontx2-af: Add mbox messages to install and delete MCAM rules Naveen Mamindlapalli
@ 2020-11-12 21:46 ` Alexander Duyck
0 siblings, 0 replies; 24+ messages in thread
From: Alexander Duyck @ 2020-11-12 21:46 UTC (permalink / raw)
To: Naveen Mamindlapalli
Cc: Netdev, LKML, Jakub Kicinski, David Miller, saeed, sgoutham,
lcherian, gakula, jerinj, sbhatta, hkelam
On Tue, Nov 10, 2020 at 11:22 PM Naveen Mamindlapalli
<naveenm@marvell.com> wrote:
>
> From: Subbaraya Sundeep <sbhatta@marvell.com>
>
> Added new mailbox messages to install and delete MCAM rules.
> These mailbox messages will be used for adding/deleting ethtool
> n-tuple filters by NIX PF. The installed MCAM rules are stored
> in a list that will be traversed later to delete the MCAM entries
> when the interface is brought down or when PCIe FLR is received.
> The delete mailbox supports deleting a single MCAM entry or range
> of entries or all the MCAM entries owned by the pcifunc. Each MCAM
> entry can be associated with a HW match stat entry if the mailbox
> requester wants to check the hit count for debugging.
>
> Modified adding default unicast DMAC match rule using install
> flow API. The default unicast DMAC match entry installed by
> Administrative Function is saved and can be changed later by the
> mailbox user to fit additional fields, or the default MCAM entry
> rule action can be used for other flow rules installed later.
>
> Modified rvu_mbox_handler_nix_lf_free mailbox to add a flag to
> disable or delete the MCAM entries. The MCAM entries are disabled
> when the interface is brought down and deleted in FLR handler.
> The disabled MCAM entries will be re-enabled when the interface
> is brought up again.
>
> Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
> Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
> Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
A couple minor issues to address, called out in comments below.
> ---
> drivers/net/ethernet/marvell/octeontx2/af/common.h | 2 +
> drivers/net/ethernet/marvell/octeontx2/af/mbox.h | 76 ++-
> drivers/net/ethernet/marvell/octeontx2/af/npc.h | 57 +-
> drivers/net/ethernet/marvell/octeontx2/af/rvu.h | 13 +
> .../net/ethernet/marvell/octeontx2/af/rvu_nix.c | 19 +-
> .../net/ethernet/marvell/octeontx2/af/rvu_npc.c | 217 ++++++-
> .../net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c | 721 +++++++++++++++++++++
> .../net/ethernet/marvell/octeontx2/nic/otx2_pf.c | 12 +-
> 8 files changed, 1065 insertions(+), 52 deletions(-)
>
<snip>
> diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
> index eb4eaa7ece3a..a7759ecfa586 100644
> --- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
> +++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
> @@ -219,7 +219,7 @@ static int npc_get_nixlf_mcam_index(struct npc_mcam *mcam,
> return npc_get_ucast_mcam_index(mcam, pcifunc, nixlf);
> }
>
> -static int npc_get_bank(struct npc_mcam *mcam, int index)
> +int npc_get_bank(struct npc_mcam *mcam, int index)
> {
> int bank = index / mcam->banksize;
>
> @@ -241,8 +241,8 @@ static bool is_mcam_entry_enabled(struct rvu *rvu, struct npc_mcam *mcam,
> return (cfg & 1);
> }
>
> -static void npc_enable_mcam_entry(struct rvu *rvu, struct npc_mcam *mcam,
> - int blkaddr, int index, bool enable)
> +void npc_enable_mcam_entry(struct rvu *rvu, struct npc_mcam *mcam,
> + int blkaddr, int index, bool enable)
> {
> int bank = npc_get_bank(mcam, index);
> int actbank = bank;
> @@ -359,6 +359,41 @@ static void npc_get_keyword(struct mcam_entry *entry, int idx,
> *cam0 = ~*cam1 & kw_mask;
> }
>
> +static void npc_get_default_entry_action(struct rvu *rvu, struct npc_mcam *mcam,
> + int blkaddr, int index,
> + struct mcam_entry *entry)
> +{
> + u16 owner, target_func;
> + struct rvu_pfvf *pfvf;
> + int bank, nixlf;
> + u64 rx_action;
> +
> + owner = mcam->entry2pfvf_map[index];
> + target_func = (entry->action >> 4) & 0xffff;
> + /* return incase target is PF or LBK or rule owner is not PF */
> + if (is_afvf(target_func) || (owner & RVU_PFVF_FUNC_MASK) ||
> + !(target_func & RVU_PFVF_FUNC_MASK))
> + return;
> +
> + pfvf = rvu_get_pfvf(rvu, target_func);
> + mcam->entry2target_pffunc[index] = target_func;
> + /* return if nixlf is not attached or initialized */
> + if (!is_nixlf_attached(rvu, target_func) || !pfvf->def_ucast_rule)
> + return;
> +
> + /* get VF ucast entry rule */
> + nix_get_nixlf(rvu, target_func, &nixlf, NULL);
> + index = npc_get_nixlf_mcam_index(mcam, target_func,
> + nixlf, NIXLF_UCAST_ENTRY);
> + bank = npc_get_bank(mcam, index);
> + index &= (mcam->banksize - 1);
> +
> + rx_action = rvu_read64(rvu, blkaddr,
> + NPC_AF_MCAMEX_BANKX_ACTION(index, bank));
> + if (rx_action)
> + entry->action = rx_action;
> +}
> +
> static void npc_config_mcam_entry(struct rvu *rvu, struct npc_mcam *mcam,
> int blkaddr, int index, u8 intf,
> struct mcam_entry *entry, bool enable)
> @@ -406,6 +441,11 @@ static void npc_config_mcam_entry(struct rvu *rvu, struct npc_mcam *mcam,
> NPC_AF_MCAMEX_BANKX_CAMX_W1(index, bank, 0), cam0);
> }
>
> + /* copy VF default entry action to the VF mcam entry */
> + if (intf == NIX_INTF_RX && actindex < mcam->bmap_entries)
> + npc_get_default_entry_action(rvu, mcam, blkaddr, actindex,
> + entry);
> +
> /* Set 'action' */
> rvu_write64(rvu, blkaddr,
> NPC_AF_MCAMEX_BANKX_ACTION(index, actbank), entry->action);
> @@ -473,11 +513,12 @@ void rvu_npc_install_ucast_entry(struct rvu *rvu, u16 pcifunc,
> int nixlf, u64 chan, u8 *mac_addr)
> {
> struct rvu_pfvf *pfvf = rvu_get_pfvf(rvu, pcifunc);
> + u8 mac_mask[] = { [0 ... ETH_ALEN] = 0xFF };
Is this supposed to be a 7 byte long array? I assume that is what is
meant by the 0 ... ETH_ALEN which would imply it covers entries 0 - 6.
This might be better as:
u8 mac_mask[ETH_ALEN] = { 0xFF, 0xFF, 0xFF, 0xFF, 0xFF, 0xFF };
> + struct npc_install_flow_req req = { 0 };
> + struct npc_install_flow_rsp rsp = { 0 };
> struct npc_mcam *mcam = &rvu->hw->mcam;
Since the mcam line is longer normally it should be before the 2 new
lines in order to maintain the reverse xmas tree format.
> - struct mcam_entry entry = { {0} };
> struct nix_rx_action action;
> - int blkaddr, index, kwi;
> - u64 mac = 0;
> + int blkaddr, index;
>
> /* AF's VFs work in promiscuous mode */
> if (is_afvf(pcifunc))
> @@ -487,20 +528,9 @@ void rvu_npc_install_ucast_entry(struct rvu *rvu, u16 pcifunc,
> if (blkaddr < 0)
> return;
>
> - for (index = ETH_ALEN - 1; index >= 0; index--)
> - mac |= ((u64)*mac_addr++) << (8 * index);
> -
> index = npc_get_nixlf_mcam_index(mcam, pcifunc,
> nixlf, NIXLF_UCAST_ENTRY);
>
> - /* Match ingress channel and DMAC */
> - entry.kw[0] = chan;
> - entry.kw_mask[0] = 0xFFFULL;
> -
> - kwi = NPC_PARSE_RESULT_DMAC_OFFSET / sizeof(u64);
> - entry.kw[kwi] = mac;
> - entry.kw_mask[kwi] = BIT_ULL(48) - 1;
> -
> /* Don't change the action if entry is already enabled
> * Otherwise RSS action may get overwritten.
> */
> @@ -513,20 +543,20 @@ void rvu_npc_install_ucast_entry(struct rvu *rvu, u16 pcifunc,
> action.pf_func = pcifunc;
> }
>
> - entry.action = *(u64 *)&action;
> - npc_config_mcam_entry(rvu, mcam, blkaddr, index,
> - pfvf->nix_rx_intf, &entry, true);
> -
> - /* add VLAN matching, setup action and save entry back for later */
> - entry.kw[0] |= (NPC_LT_LB_STAG_QINQ | NPC_LT_LB_CTAG) << 20;
> - entry.kw_mask[0] |= (NPC_LT_LB_STAG_QINQ & NPC_LT_LB_CTAG) << 20;
> + req.default_rule = 1;
> + ether_addr_copy(req.packet.dmac, mac_addr);
> + ether_addr_copy(req.mask.dmac, mac_mask);
If this is all you were using the mac_mask for you could probably just
use a memset here to achieve the same thing and save yourself the
trouble of allocating the mac_mask. See eth_broadcast_addr().
> + req.features = BIT_ULL(NPC_DMAC);
> + req.channel = chan;
> + req.intf = pfvf->nix_rx_intf;
> + req.op = action.op;
> + req.hdr.pcifunc = 0; /* AF is requester */
> + req.vf = action.pf_func;
> + req.index = action.index;
> + req.match_id = action.match_id;
> + req.flow_key_alg = action.flow_key_alg;
>
> - entry.vtag_action = VTAG0_VALID_BIT |
> - FIELD_PREP(VTAG0_TYPE_MASK, 0) |
> - FIELD_PREP(VTAG0_LID_MASK, NPC_LID_LA) |
> - FIELD_PREP(VTAG0_RELPTR_MASK, 12);
> -
> - memcpy(&pfvf->entry, &entry, sizeof(entry));
> + rvu_mbox_handler_npc_install_flow(rvu, &req, &rsp);
> }
>
<snip>
> +static void npc_update_flow(struct rvu *rvu, struct mcam_entry *entry,
> + u64 features, struct flow_msg *pkt,
> + struct flow_msg *mask,
> + struct rvu_npc_mcam_rule *output, u8 intf)
> +{
> + u64 dmac_mask = ether_addr_to_u64(mask->dmac);
> + u64 smac_mask = ether_addr_to_u64(mask->smac);
> + u64 dmac_val = ether_addr_to_u64(pkt->dmac);
> + u64 smac_val = ether_addr_to_u64(pkt->smac);
> + struct flow_msg *opkt = &output->packet;
> + struct flow_msg *omask = &output->mask;
> +
> + if (!features)
> + return;
> +
> +#define NPC_WRITE_FLOW(field, member, val_lo, val_hi, mask_lo, mask_hi) \
> +do { \
> + if (features & BIT_ULL((field))) { \
> + npc_update_entry(rvu, (field), entry, (val_lo), (val_hi), \
> + (mask_lo), (mask_hi), intf); \
> + memcpy(&opkt->member, &pkt->member, sizeof(pkt->member)); \
> + memcpy(&omask->member, &mask->member, sizeof(mask->member)); \
> + } \
> +} while (0)
> +
The placement of this macro seems wierd to me. Either it needs to be
boe moved before the function declaration, or it should be defined
where it is used instead of this block above it.
> + /* For tcp/udp/sctp LTYPE should be present in entry */
> + if (features & (BIT_ULL(NPC_SPORT_TCP) | BIT_ULL(NPC_DPORT_TCP)))
> + npc_update_entry(rvu, NPC_LD, entry, NPC_LT_LD_TCP,
> + 0, ~0ULL, 0, intf);
> + if (features & (BIT_ULL(NPC_SPORT_UDP) | BIT_ULL(NPC_DPORT_UDP)))
> + npc_update_entry(rvu, NPC_LD, entry, NPC_LT_LD_UDP,
> + 0, ~0ULL, 0, intf);
> + if (features & (BIT_ULL(NPC_SPORT_SCTP) | BIT_ULL(NPC_DPORT_SCTP)))
> + npc_update_entry(rvu, NPC_LD, entry, NPC_LT_LD_SCTP,
> + 0, ~0ULL, 0, intf);
> +
> + if (features & BIT_ULL(NPC_OUTER_VID))
> + npc_update_entry(rvu, NPC_LB, entry,
> + NPC_LT_LB_STAG_QINQ | NPC_LT_LB_CTAG, 0,
> + NPC_LT_LB_STAG_QINQ & NPC_LT_LB_CTAG, 0, intf);
> +
> + NPC_WRITE_FLOW(NPC_DMAC, dmac, dmac_val, 0, dmac_mask, 0);
> + NPC_WRITE_FLOW(NPC_SMAC, smac, smac_val, 0, smac_mask, 0);
> + NPC_WRITE_FLOW(NPC_ETYPE, etype, ntohs(pkt->etype), 0,
> + ntohs(mask->etype), 0);
> + NPC_WRITE_FLOW(NPC_SIP_IPV4, ip4src, ntohl(pkt->ip4src), 0,
> + ntohl(mask->ip4src), 0);
> + NPC_WRITE_FLOW(NPC_DIP_IPV4, ip4dst, ntohl(pkt->ip4dst), 0,
> + ntohl(mask->ip4dst), 0);
> + NPC_WRITE_FLOW(NPC_SPORT_TCP, sport, ntohs(pkt->sport), 0,
> + ntohs(mask->sport), 0);
> + NPC_WRITE_FLOW(NPC_SPORT_UDP, sport, ntohs(pkt->sport), 0,
> + ntohs(mask->sport), 0);
> + NPC_WRITE_FLOW(NPC_DPORT_TCP, dport, ntohs(pkt->dport), 0,
> + ntohs(mask->dport), 0);
> + NPC_WRITE_FLOW(NPC_DPORT_UDP, dport, ntohs(pkt->dport), 0,
> + ntohs(mask->dport), 0);
> + NPC_WRITE_FLOW(NPC_SPORT_SCTP, sport, ntohs(pkt->sport), 0,
> + ntohs(mask->sport), 0);
> + NPC_WRITE_FLOW(NPC_DPORT_SCTP, dport, ntohs(pkt->dport), 0,
> + ntohs(mask->dport), 0);
> +
> + NPC_WRITE_FLOW(NPC_OUTER_VID, vlan_tci, ntohs(pkt->vlan_tci), 0,
> + ntohs(mask->vlan_tci), 0);
> +
> + npc_update_ipv6_flow(rvu, entry, features, pkt, mask, output, intf);
> +}
> +
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH v3 net-next 05/13] octeontx2-pf: Add support for ethtool ntuple filters
2020-11-11 7:13 [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Naveen Mamindlapalli
` (3 preceding siblings ...)
2020-11-11 7:13 ` [PATCH v3 net-next 04/13] octeontx2-af: Add mbox messages to install and delete MCAM rules Naveen Mamindlapalli
@ 2020-11-11 7:13 ` Naveen Mamindlapalli
2020-11-11 7:13 ` [PATCH v3 net-next 06/13] octeontx2-pf: Add support for unicast MAC address filtering Naveen Mamindlapalli
` (8 subsequent siblings)
13 siblings, 0 replies; 24+ messages in thread
From: Naveen Mamindlapalli @ 2020-11-11 7:13 UTC (permalink / raw)
To: netdev, linux-kernel
Cc: kuba, davem, saeed, sgoutham, lcherian, gakula, jerinj, sbhatta,
hkelam
From: Subbaraya Sundeep <sbhatta@marvell.com>
This patch adds support for adding and deleting ethtool ntuple
filters. The filters for ether, ipv4, ipv6, tcp, udp and sctp
are supported. The mask is also supported. The supported actions
are drop and direct to a queue. Additionally we support FLOW_EXT
field vlan_tci and FLOW_MAC_EXT.
The NIX PF will allocate total 32 MCAM entries for the use of
ethtool ntuple filters. The Administrative Function(AF) will
install/delete the MCAM rules when NIX PF sends mailbox message
to install/delete the ntuple filters.
Ethtool ntuple filters support is restricted to PFs as of now
and PF can install ntuple filters to direct the traffic to its
VFs. Hence added a separate callback for VFs to get/set RSS
configuration.
Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
---
.../net/ethernet/marvell/octeontx2/nic/Makefile | 2 +-
.../ethernet/marvell/octeontx2/nic/otx2_common.h | 31 ++
.../ethernet/marvell/octeontx2/nic/otx2_ethtool.c | 58 +-
.../ethernet/marvell/octeontx2/nic/otx2_flows.c | 603 +++++++++++++++++++++
.../net/ethernet/marvell/octeontx2/nic/otx2_pf.c | 18 +-
5 files changed, 707 insertions(+), 5 deletions(-)
create mode 100644 drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
index b2c6385707c9..4193ae3bde6b 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/Makefile
@@ -7,7 +7,7 @@ obj-$(CONFIG_OCTEONTX2_PF) += octeontx2_nicpf.o
obj-$(CONFIG_OCTEONTX2_VF) += octeontx2_nicvf.o
octeontx2_nicpf-y := otx2_pf.o otx2_common.o otx2_txrx.o otx2_ethtool.o \
- otx2_ptp.o
+ otx2_ptp.o otx2_flows.o
octeontx2_nicvf-y := otx2_vf.o
ccflags-y += -I$(srctree)/drivers/net/ethernet/marvell/octeontx2/af
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index 386cb08497e4..2387c40a2a8f 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -18,6 +18,7 @@
#include <linux/timecounter.h>
#include <mbox.h>
+#include <npc.h>
#include "otx2_reg.h"
#include "otx2_txrx.h"
#include <rvu_trace.h>
@@ -228,6 +229,16 @@ struct otx2_ptp {
#define OTX2_HW_TIMESTAMP_LEN 8
+struct otx2_flow_config {
+ u16 entry[NPC_MAX_NONCONTIG_ENTRIES];
+ u32 nr_flows;
+#define OTX2_MAX_NTUPLE_FLOWS 32
+#define OTX2_MCAM_COUNT OTX2_MAX_NTUPLE_FLOWS
+ u32 ntuple_offset;
+ u32 ntuple_max_flows;
+ struct list_head flow_list;
+};
+
struct otx2_nic {
void __iomem *reg_base;
struct net_device *netdev;
@@ -238,6 +249,8 @@ struct otx2_nic {
#define OTX2_FLAG_RX_TSTAMP_ENABLED BIT_ULL(0)
#define OTX2_FLAG_TX_TSTAMP_ENABLED BIT_ULL(1)
#define OTX2_FLAG_INTF_DOWN BIT_ULL(2)
+#define OTX2_FLAG_MCAM_ENTRIES_ALLOC BIT_ULL(3)
+#define OTX2_FLAG_NTUPLE_SUPPORT BIT_ULL(4)
#define OTX2_FLAG_RX_PAUSE_ENABLED BIT_ULL(9)
#define OTX2_FLAG_TX_PAUSE_ENABLED BIT_ULL(10)
u64 flags;
@@ -275,6 +288,8 @@ struct otx2_nic {
struct otx2_ptp *ptp;
struct hwtstamp_config tstamp;
+
+ struct otx2_flow_config *flow_cfg;
};
static inline bool is_otx2_lbkvf(struct pci_dev *pdev)
@@ -644,4 +659,20 @@ int otx2_open(struct net_device *netdev);
int otx2_stop(struct net_device *netdev);
int otx2_set_real_num_queues(struct net_device *netdev,
int tx_queues, int rx_queues);
+/* MCAM filter related APIs */
+int otx2_mcam_flow_init(struct otx2_nic *pf);
+int otx2_alloc_mcam_entries(struct otx2_nic *pfvf);
+void otx2_mcam_flow_del(struct otx2_nic *pf);
+int otx2_destroy_ntuple_flows(struct otx2_nic *pf);
+int otx2_destroy_mcam_flows(struct otx2_nic *pfvf);
+int otx2_get_flow(struct otx2_nic *pfvf,
+ struct ethtool_rxnfc *nfc, u32 location);
+int otx2_get_all_flows(struct otx2_nic *pfvf,
+ struct ethtool_rxnfc *nfc, u32 *rule_locs);
+int otx2_add_flow(struct otx2_nic *pfvf,
+ struct ethtool_rx_flow_spec *fsp);
+int otx2_remove_flow(struct otx2_nic *pfvf, u32 location);
+int otx2_prepare_flow_request(struct ethtool_rx_flow_spec *fsp,
+ struct npc_install_flow_req *req);
+
#endif /* OTX2_COMMON_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c
index 662fb80dbb9d..67171b66a56c 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_ethtool.c
@@ -551,6 +551,16 @@ static int otx2_get_rxnfc(struct net_device *dev,
nfc->data = pfvf->hw.rx_queues;
ret = 0;
break;
+ case ETHTOOL_GRXCLSRLCNT:
+ nfc->rule_cnt = pfvf->flow_cfg->nr_flows;
+ ret = 0;
+ break;
+ case ETHTOOL_GRXCLSRULE:
+ ret = otx2_get_flow(pfvf, nfc, nfc->fs.location);
+ break;
+ case ETHTOOL_GRXCLSRLALL:
+ ret = otx2_get_all_flows(pfvf, nfc, rules);
+ break;
case ETHTOOL_GRXFH:
return otx2_get_rss_hash_opts(pfvf, nfc);
default:
@@ -560,6 +570,50 @@ static int otx2_get_rxnfc(struct net_device *dev,
}
static int otx2_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *nfc)
+{
+ bool ntuple = !!(dev->features & NETIF_F_NTUPLE);
+ struct otx2_nic *pfvf = netdev_priv(dev);
+ int ret = -EOPNOTSUPP;
+
+ switch (nfc->cmd) {
+ case ETHTOOL_SRXFH:
+ ret = otx2_set_rss_hash_opts(pfvf, nfc);
+ break;
+ case ETHTOOL_SRXCLSRLINS:
+ if (netif_running(dev) && ntuple)
+ ret = otx2_add_flow(pfvf, &nfc->fs);
+ break;
+ case ETHTOOL_SRXCLSRLDEL:
+ if (netif_running(dev) && ntuple)
+ ret = otx2_remove_flow(pfvf, nfc->fs.location);
+ break;
+ default:
+ break;
+ }
+
+ return ret;
+}
+
+static int otx2vf_get_rxnfc(struct net_device *dev,
+ struct ethtool_rxnfc *nfc, u32 *rules)
+{
+ struct otx2_nic *pfvf = netdev_priv(dev);
+ int ret = -EOPNOTSUPP;
+
+ switch (nfc->cmd) {
+ case ETHTOOL_GRXRINGS:
+ nfc->data = pfvf->hw.rx_queues;
+ ret = 0;
+ break;
+ case ETHTOOL_GRXFH:
+ return otx2_get_rss_hash_opts(pfvf, nfc);
+ default:
+ break;
+ }
+ return ret;
+}
+
+static int otx2vf_set_rxnfc(struct net_device *dev, struct ethtool_rxnfc *nfc)
{
struct otx2_nic *pfvf = netdev_priv(dev);
int ret = -EOPNOTSUPP;
@@ -806,8 +860,8 @@ static const struct ethtool_ops otx2vf_ethtool_ops = {
.get_sset_count = otx2vf_get_sset_count,
.set_channels = otx2_set_channels,
.get_channels = otx2_get_channels,
- .get_rxnfc = otx2_get_rxnfc,
- .set_rxnfc = otx2_set_rxnfc,
+ .get_rxnfc = otx2vf_get_rxnfc,
+ .set_rxnfc = otx2vf_set_rxnfc,
.get_rxfh_key_size = otx2_get_rxfh_key_size,
.get_rxfh_indir_size = otx2_get_rxfh_indir_size,
.get_rxfh = otx2_get_rxfh,
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c
new file mode 100644
index 000000000000..094f547c9889
--- /dev/null
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c
@@ -0,0 +1,603 @@
+// SPDX-License-Identifier: GPL-2.0
+/* Marvell OcteonTx2 RVU Physcial Function ethernet driver
+ *
+ * Copyright (C) 2020 Marvell International Ltd.
+ */
+
+#include <net/ipv6.h>
+
+#include "otx2_common.h"
+
+#define OTX2_DEFAULT_ACTION 0x1
+
+struct otx2_flow {
+ struct ethtool_rx_flow_spec flow_spec;
+ struct list_head list;
+ u32 location;
+ u16 entry;
+ bool is_vf;
+ int vf;
+};
+
+int otx2_alloc_mcam_entries(struct otx2_nic *pfvf)
+{
+ struct otx2_flow_config *flow_cfg = pfvf->flow_cfg;
+ struct npc_mcam_alloc_entry_req *req;
+ struct npc_mcam_alloc_entry_rsp *rsp;
+ int i;
+
+ mutex_lock(&pfvf->mbox.lock);
+
+ req = otx2_mbox_alloc_msg_npc_mcam_alloc_entry(&pfvf->mbox);
+ if (!req) {
+ mutex_unlock(&pfvf->mbox.lock);
+ return -ENOMEM;
+ }
+
+ req->contig = false;
+ req->count = OTX2_MCAM_COUNT;
+
+ /* Send message to AF */
+ if (otx2_sync_mbox_msg(&pfvf->mbox)) {
+ mutex_unlock(&pfvf->mbox.lock);
+ return -EINVAL;
+ }
+
+ rsp = (struct npc_mcam_alloc_entry_rsp *)otx2_mbox_get_rsp
+ (&pfvf->mbox.mbox, 0, &req->hdr);
+
+ if (rsp->count != req->count)
+ netdev_info(pfvf->netdev,
+ "Unable to allocate %d MCAM entries, got %d\n",
+ req->count, rsp->count);
+
+ flow_cfg->ntuple_max_flows = rsp->count;
+ flow_cfg->ntuple_offset = 0;
+ pfvf->flags |= OTX2_FLAG_NTUPLE_SUPPORT;
+
+ for (i = 0; i < rsp->count; i++)
+ flow_cfg->entry[i] = rsp->entry_list[i];
+
+ pfvf->flags |= OTX2_FLAG_MCAM_ENTRIES_ALLOC;
+
+ mutex_unlock(&pfvf->mbox.lock);
+
+ return 0;
+}
+
+int otx2_mcam_flow_init(struct otx2_nic *pf)
+{
+ int err;
+
+ pf->flow_cfg = devm_kzalloc(pf->dev, sizeof(struct otx2_flow_config),
+ GFP_KERNEL);
+ if (!pf->flow_cfg)
+ return -ENOMEM;
+
+ INIT_LIST_HEAD(&pf->flow_cfg->flow_list);
+
+ pf->flow_cfg->ntuple_max_flows = OTX2_MAX_NTUPLE_FLOWS;
+
+ err = otx2_alloc_mcam_entries(pf);
+ if (err)
+ return err;
+
+ return 0;
+}
+
+void otx2_mcam_flow_del(struct otx2_nic *pf)
+{
+ otx2_destroy_mcam_flows(pf);
+}
+
+static struct otx2_flow *otx2_find_flow(struct otx2_nic *pfvf, u32 location)
+{
+ struct otx2_flow *iter;
+
+ list_for_each_entry(iter, &pfvf->flow_cfg->flow_list, list) {
+ if (iter->location == location)
+ return iter;
+ }
+
+ return NULL;
+}
+
+static void otx2_add_flow_to_list(struct otx2_nic *pfvf, struct otx2_flow *flow)
+{
+ struct list_head *head = &pfvf->flow_cfg->flow_list;
+ struct otx2_flow *iter;
+
+ list_for_each_entry(iter, &pfvf->flow_cfg->flow_list, list) {
+ if (iter->location > flow->location)
+ break;
+ head = &iter->list;
+ }
+
+ list_add(&flow->list, head);
+}
+
+int otx2_get_flow(struct otx2_nic *pfvf, struct ethtool_rxnfc *nfc,
+ u32 location)
+{
+ struct otx2_flow *iter;
+
+ if (location >= pfvf->flow_cfg->ntuple_max_flows)
+ return -EINVAL;
+
+ list_for_each_entry(iter, &pfvf->flow_cfg->flow_list, list) {
+ if (iter->location == location) {
+ nfc->fs = iter->flow_spec;
+ return 0;
+ }
+ }
+
+ return -ENOENT;
+}
+
+int otx2_get_all_flows(struct otx2_nic *pfvf, struct ethtool_rxnfc *nfc,
+ u32 *rule_locs)
+{
+ u32 location = 0;
+ int idx = 0;
+ int err = 0;
+
+ nfc->data = pfvf->flow_cfg->ntuple_max_flows;
+ while ((!err || err == -ENOENT) && idx < nfc->rule_cnt) {
+ err = otx2_get_flow(pfvf, nfc, location);
+ if (!err)
+ rule_locs[idx++] = location;
+ location++;
+ }
+
+ return err;
+}
+
+static void otx2_prepare_ipv4_flow(struct ethtool_rx_flow_spec *fsp,
+ struct npc_install_flow_req *req,
+ u32 flow_type)
+{
+ struct ethtool_usrip4_spec *ipv4_usr_mask = &fsp->m_u.usr_ip4_spec;
+ struct ethtool_usrip4_spec *ipv4_usr_hdr = &fsp->h_u.usr_ip4_spec;
+ struct ethtool_tcpip4_spec *ipv4_l4_mask = &fsp->m_u.tcp_ip4_spec;
+ struct ethtool_tcpip4_spec *ipv4_l4_hdr = &fsp->h_u.tcp_ip4_spec;
+ struct flow_msg *pmask = &req->mask;
+ struct flow_msg *pkt = &req->packet;
+
+ switch (flow_type) {
+ case IP_USER_FLOW:
+ if (ipv4_usr_mask->ip4src) {
+ memcpy(&pkt->ip4src, &ipv4_usr_hdr->ip4src,
+ sizeof(pkt->ip4src));
+ memcpy(&pmask->ip4src, &ipv4_usr_mask->ip4src,
+ sizeof(pmask->ip4src));
+ req->features |= BIT_ULL(NPC_SIP_IPV4);
+ }
+ if (ipv4_usr_mask->ip4dst) {
+ memcpy(&pkt->ip4dst, &ipv4_usr_hdr->ip4dst,
+ sizeof(pkt->ip4dst));
+ memcpy(&pmask->ip4dst, &ipv4_usr_mask->ip4dst,
+ sizeof(pmask->ip4dst));
+ req->features |= BIT_ULL(NPC_DIP_IPV4);
+ }
+ break;
+ case TCP_V4_FLOW:
+ case UDP_V4_FLOW:
+ case SCTP_V4_FLOW:
+ if (ipv4_l4_mask->ip4src) {
+ memcpy(&pkt->ip4src, &ipv4_l4_hdr->ip4src,
+ sizeof(pkt->ip4src));
+ memcpy(&pmask->ip4src, &ipv4_l4_mask->ip4src,
+ sizeof(pmask->ip4src));
+ req->features |= BIT_ULL(NPC_SIP_IPV4);
+ }
+ if (ipv4_l4_mask->ip4dst) {
+ memcpy(&pkt->ip4dst, &ipv4_l4_hdr->ip4dst,
+ sizeof(pkt->ip4dst));
+ memcpy(&pmask->ip4dst, &ipv4_l4_mask->ip4dst,
+ sizeof(pmask->ip4dst));
+ req->features |= BIT_ULL(NPC_DIP_IPV4);
+ }
+ if (ipv4_l4_mask->psrc) {
+ memcpy(&pkt->sport, &ipv4_l4_hdr->psrc,
+ sizeof(pkt->sport));
+ memcpy(&pmask->sport, &ipv4_l4_mask->psrc,
+ sizeof(pmask->sport));
+ if (flow_type == UDP_V4_FLOW)
+ req->features |= BIT_ULL(NPC_SPORT_UDP);
+ else if (flow_type == TCP_V4_FLOW)
+ req->features |= BIT_ULL(NPC_SPORT_TCP);
+ else
+ req->features |= BIT_ULL(NPC_SPORT_SCTP);
+ }
+ if (ipv4_l4_mask->pdst) {
+ memcpy(&pkt->dport, &ipv4_l4_hdr->pdst,
+ sizeof(pkt->dport));
+ memcpy(&pmask->dport, &ipv4_l4_mask->pdst,
+ sizeof(pmask->dport));
+ if (flow_type == UDP_V4_FLOW)
+ req->features |= BIT_ULL(NPC_DPORT_UDP);
+ else if (flow_type == TCP_V4_FLOW)
+ req->features |= BIT_ULL(NPC_DPORT_TCP);
+ else
+ req->features |= BIT_ULL(NPC_DPORT_SCTP);
+ }
+ break;
+ default:
+ break;
+ }
+}
+
+static void otx2_prepare_ipv6_flow(struct ethtool_rx_flow_spec *fsp,
+ struct npc_install_flow_req *req,
+ u32 flow_type)
+{
+ struct ethtool_usrip6_spec *ipv6_usr_mask = &fsp->m_u.usr_ip6_spec;
+ struct ethtool_usrip6_spec *ipv6_usr_hdr = &fsp->h_u.usr_ip6_spec;
+ struct ethtool_tcpip6_spec *ipv6_l4_mask = &fsp->m_u.tcp_ip6_spec;
+ struct ethtool_tcpip6_spec *ipv6_l4_hdr = &fsp->h_u.tcp_ip6_spec;
+ struct flow_msg *pmask = &req->mask;
+ struct flow_msg *pkt = &req->packet;
+
+ switch (flow_type) {
+ case IPV6_USER_FLOW:
+ if (!ipv6_addr_any((struct in6_addr *)ipv6_usr_mask->ip6src)) {
+ memcpy(&pkt->ip6src, &ipv6_usr_hdr->ip6src,
+ sizeof(pkt->ip6src));
+ memcpy(&pmask->ip6src, &ipv6_usr_mask->ip6src,
+ sizeof(pmask->ip6src));
+ req->features |= BIT_ULL(NPC_SIP_IPV6);
+ }
+ if (!ipv6_addr_any((struct in6_addr *)ipv6_usr_mask->ip6dst)) {
+ memcpy(&pkt->ip6dst, &ipv6_usr_hdr->ip6dst,
+ sizeof(pkt->ip6dst));
+ memcpy(&pmask->ip6dst, &ipv6_usr_mask->ip6dst,
+ sizeof(pmask->ip6dst));
+ req->features |= BIT_ULL(NPC_DIP_IPV6);
+ }
+ break;
+ case TCP_V6_FLOW:
+ case UDP_V6_FLOW:
+ case SCTP_V6_FLOW:
+ if (!ipv6_addr_any((struct in6_addr *)ipv6_l4_mask->ip6src)) {
+ memcpy(&pkt->ip6src, &ipv6_l4_hdr->ip6src,
+ sizeof(pkt->ip6src));
+ memcpy(&pmask->ip6src, &ipv6_l4_mask->ip6src,
+ sizeof(pmask->ip6src));
+ req->features |= BIT_ULL(NPC_SIP_IPV6);
+ }
+ if (!ipv6_addr_any((struct in6_addr *)ipv6_l4_mask->ip6dst)) {
+ memcpy(&pkt->ip6dst, &ipv6_l4_hdr->ip6dst,
+ sizeof(pkt->ip6dst));
+ memcpy(&pmask->ip6dst, &ipv6_l4_mask->ip6dst,
+ sizeof(pmask->ip6dst));
+ req->features |= BIT_ULL(NPC_DIP_IPV6);
+ }
+ if (ipv6_l4_mask->psrc) {
+ memcpy(&pkt->sport, &ipv6_l4_hdr->psrc,
+ sizeof(pkt->sport));
+ memcpy(&pmask->sport, &ipv6_l4_mask->psrc,
+ sizeof(pmask->sport));
+ if (flow_type == UDP_V6_FLOW)
+ req->features |= BIT_ULL(NPC_SPORT_UDP);
+ else if (flow_type == TCP_V6_FLOW)
+ req->features |= BIT_ULL(NPC_SPORT_TCP);
+ else
+ req->features |= BIT_ULL(NPC_SPORT_SCTP);
+ }
+ if (ipv6_l4_mask->pdst) {
+ memcpy(&pkt->dport, &ipv6_l4_hdr->pdst,
+ sizeof(pkt->dport));
+ memcpy(&pmask->dport, &ipv6_l4_mask->pdst,
+ sizeof(pmask->dport));
+ if (flow_type == UDP_V6_FLOW)
+ req->features |= BIT_ULL(NPC_DPORT_UDP);
+ else if (flow_type == TCP_V6_FLOW)
+ req->features |= BIT_ULL(NPC_DPORT_TCP);
+ else
+ req->features |= BIT_ULL(NPC_DPORT_SCTP);
+ }
+ break;
+ default:
+ break;
+ }
+}
+
+int otx2_prepare_flow_request(struct ethtool_rx_flow_spec *fsp,
+ struct npc_install_flow_req *req)
+{
+ struct ethhdr *eth_mask = &fsp->m_u.ether_spec;
+ struct ethhdr *eth_hdr = &fsp->h_u.ether_spec;
+ struct flow_msg *pmask = &req->mask;
+ struct flow_msg *pkt = &req->packet;
+ u32 flow_type;
+
+ flow_type = fsp->flow_type & ~(FLOW_EXT | FLOW_MAC_EXT);
+ switch (flow_type) {
+ /* bits not set in mask are don't care */
+ case ETHER_FLOW:
+ if (!is_zero_ether_addr(eth_mask->h_source)) {
+ ether_addr_copy(pkt->smac, eth_hdr->h_source);
+ ether_addr_copy(pmask->smac, eth_mask->h_source);
+ req->features |= BIT_ULL(NPC_SMAC);
+ }
+ if (!is_zero_ether_addr(eth_mask->h_dest)) {
+ ether_addr_copy(pkt->dmac, eth_hdr->h_dest);
+ ether_addr_copy(pmask->dmac, eth_mask->h_dest);
+ req->features |= BIT_ULL(NPC_DMAC);
+ }
+ if (eth_mask->h_proto) {
+ memcpy(&pkt->etype, ð_hdr->h_proto,
+ sizeof(pkt->etype));
+ memcpy(&pmask->etype, ð_mask->h_proto,
+ sizeof(pmask->etype));
+ req->features |= BIT_ULL(NPC_ETYPE);
+ }
+ break;
+ case IP_USER_FLOW:
+ case TCP_V4_FLOW:
+ case UDP_V4_FLOW:
+ case SCTP_V4_FLOW:
+ otx2_prepare_ipv4_flow(fsp, req, flow_type);
+ break;
+ case IPV6_USER_FLOW:
+ case TCP_V6_FLOW:
+ case UDP_V6_FLOW:
+ case SCTP_V6_FLOW:
+ otx2_prepare_ipv6_flow(fsp, req, flow_type);
+ break;
+ default:
+ return -EOPNOTSUPP;
+ }
+ if (fsp->flow_type & FLOW_EXT) {
+ if (fsp->m_ext.vlan_etype)
+ return -EINVAL;
+ if (fsp->m_ext.vlan_tci) {
+ if (fsp->m_ext.vlan_tci != cpu_to_be16(VLAN_VID_MASK))
+ return -EINVAL;
+ if (be16_to_cpu(fsp->h_ext.vlan_tci) >= VLAN_N_VID)
+ return -EINVAL;
+
+ memcpy(&pkt->vlan_tci, &fsp->h_ext.vlan_tci,
+ sizeof(pkt->vlan_tci));
+ memcpy(&pmask->vlan_tci, &fsp->m_ext.vlan_tci,
+ sizeof(pmask->vlan_tci));
+ req->features |= BIT_ULL(NPC_OUTER_VID);
+ }
+
+ /* Not Drop/Direct to queue but use action in default entry */
+ if (fsp->m_ext.data[1] &&
+ fsp->h_ext.data[1] == cpu_to_be32(OTX2_DEFAULT_ACTION))
+ req->op = NIX_RX_ACTION_DEFAULT;
+ }
+
+ if (fsp->flow_type & FLOW_MAC_EXT &&
+ !is_zero_ether_addr(fsp->m_ext.h_dest)) {
+ ether_addr_copy(pkt->dmac, fsp->h_ext.h_dest);
+ ether_addr_copy(pmask->dmac, fsp->m_ext.h_dest);
+ req->features |= BIT_ULL(NPC_DMAC);
+ }
+
+ if (!req->features)
+ return -EOPNOTSUPP;
+
+ return 0;
+}
+
+static int otx2_add_flow_msg(struct otx2_nic *pfvf, struct otx2_flow *flow)
+{
+ u64 ring_cookie = flow->flow_spec.ring_cookie;
+ struct npc_install_flow_req *req;
+ int err, vf = 0;
+
+ mutex_lock(&pfvf->mbox.lock);
+ req = otx2_mbox_alloc_msg_npc_install_flow(&pfvf->mbox);
+ if (!req) {
+ mutex_unlock(&pfvf->mbox.lock);
+ return -ENOMEM;
+ }
+
+ err = otx2_prepare_flow_request(&flow->flow_spec, req);
+ if (err) {
+ /* free the allocated msg above */
+ otx2_mbox_reset(&pfvf->mbox.mbox, 0);
+ mutex_unlock(&pfvf->mbox.lock);
+ return err;
+ }
+
+ req->entry = flow->entry;
+ req->intf = NIX_INTF_RX;
+ req->set_cntr = 1;
+ req->channel = pfvf->hw.rx_chan_base;
+ if (ring_cookie == RX_CLS_FLOW_DISC) {
+ req->op = NIX_RX_ACTIONOP_DROP;
+ } else {
+ /* change to unicast only if action of default entry is not
+ * requested by user
+ */
+ if (req->op != NIX_RX_ACTION_DEFAULT)
+ req->op = NIX_RX_ACTIONOP_UCAST;
+ req->index = ethtool_get_flow_spec_ring(ring_cookie);
+ vf = ethtool_get_flow_spec_ring_vf(ring_cookie);
+ if (vf > pci_num_vf(pfvf->pdev)) {
+ mutex_unlock(&pfvf->mbox.lock);
+ return -EINVAL;
+ }
+ }
+
+ /* ethtool ring_cookie has (VF + 1) for VF */
+ if (vf) {
+ req->vf = vf;
+ flow->is_vf = true;
+ flow->vf = vf;
+ }
+
+ /* Send message to AF */
+ err = otx2_sync_mbox_msg(&pfvf->mbox);
+ mutex_unlock(&pfvf->mbox.lock);
+ return err;
+}
+
+int otx2_add_flow(struct otx2_nic *pfvf, struct ethtool_rx_flow_spec *fsp)
+{
+ struct otx2_flow_config *flow_cfg = pfvf->flow_cfg;
+ u32 ring = ethtool_get_flow_spec_ring(fsp->ring_cookie);
+ struct otx2_flow *flow;
+ bool new = false;
+ int err;
+
+ if (!(pfvf->flags & OTX2_FLAG_NTUPLE_SUPPORT))
+ return -ENOMEM;
+
+ if (ring >= pfvf->hw.rx_queues && fsp->ring_cookie != RX_CLS_FLOW_DISC)
+ return -EINVAL;
+
+ if (fsp->location >= flow_cfg->ntuple_max_flows)
+ return -EINVAL;
+
+ flow = otx2_find_flow(pfvf, fsp->location);
+ if (!flow) {
+ flow = kzalloc(sizeof(*flow), GFP_ATOMIC);
+ if (!flow)
+ return -ENOMEM;
+ flow->location = fsp->location;
+ flow->entry = flow_cfg->entry[flow_cfg->ntuple_offset +
+ flow->location];
+ new = true;
+ }
+ /* struct copy */
+ flow->flow_spec = *fsp;
+
+ err = otx2_add_flow_msg(pfvf, flow);
+ if (err) {
+ if (new)
+ kfree(flow);
+ return err;
+ }
+
+ /* add the new flow installed to list */
+ if (new) {
+ otx2_add_flow_to_list(pfvf, flow);
+ flow_cfg->nr_flows++;
+ }
+
+ return 0;
+}
+
+static int otx2_remove_flow_msg(struct otx2_nic *pfvf, u16 entry, bool all)
+{
+ struct npc_delete_flow_req *req;
+ int err;
+
+ mutex_lock(&pfvf->mbox.lock);
+ req = otx2_mbox_alloc_msg_npc_delete_flow(&pfvf->mbox);
+ if (!req) {
+ mutex_unlock(&pfvf->mbox.lock);
+ return -ENOMEM;
+ }
+
+ req->entry = entry;
+ if (all)
+ req->all = 1;
+
+ /* Send message to AF */
+ err = otx2_sync_mbox_msg(&pfvf->mbox);
+ mutex_unlock(&pfvf->mbox.lock);
+ return err;
+}
+
+int otx2_remove_flow(struct otx2_nic *pfvf, u32 location)
+{
+ struct otx2_flow_config *flow_cfg = pfvf->flow_cfg;
+ struct otx2_flow *flow;
+ int err;
+
+ if (location >= flow_cfg->ntuple_max_flows)
+ return -EINVAL;
+
+ flow = otx2_find_flow(pfvf, location);
+ if (!flow)
+ return -ENOENT;
+
+ err = otx2_remove_flow_msg(pfvf, flow->entry, false);
+ if (err)
+ return err;
+
+ list_del(&flow->list);
+ kfree(flow);
+ flow_cfg->nr_flows--;
+
+ return 0;
+}
+
+int otx2_destroy_ntuple_flows(struct otx2_nic *pfvf)
+{
+ struct otx2_flow_config *flow_cfg = pfvf->flow_cfg;
+ struct npc_delete_flow_req *req;
+ struct otx2_flow *iter, *tmp;
+ int err;
+
+ if (!(pfvf->flags & OTX2_FLAG_NTUPLE_SUPPORT))
+ return 0;
+
+ mutex_lock(&pfvf->mbox.lock);
+ req = otx2_mbox_alloc_msg_npc_delete_flow(&pfvf->mbox);
+ if (!req) {
+ mutex_unlock(&pfvf->mbox.lock);
+ return -ENOMEM;
+ }
+
+ req->start = flow_cfg->entry[flow_cfg->ntuple_offset];
+ req->end = flow_cfg->entry[flow_cfg->ntuple_offset +
+ flow_cfg->ntuple_max_flows - 1];
+ err = otx2_sync_mbox_msg(&pfvf->mbox);
+ mutex_unlock(&pfvf->mbox.lock);
+
+ list_for_each_entry_safe(iter, tmp, &flow_cfg->flow_list, list) {
+ list_del(&iter->list);
+ kfree(iter);
+ flow_cfg->nr_flows--;
+ }
+ return err;
+}
+
+int otx2_destroy_mcam_flows(struct otx2_nic *pfvf)
+{
+ struct otx2_flow_config *flow_cfg = pfvf->flow_cfg;
+ struct npc_mcam_free_entry_req *req;
+ struct otx2_flow *iter, *tmp;
+ int err;
+
+ if (!(pfvf->flags & OTX2_FLAG_MCAM_ENTRIES_ALLOC))
+ return 0;
+
+ /* remove all flows */
+ err = otx2_remove_flow_msg(pfvf, 0, true);
+ if (err)
+ return err;
+
+ list_for_each_entry_safe(iter, tmp, &flow_cfg->flow_list, list) {
+ list_del(&iter->list);
+ kfree(iter);
+ flow_cfg->nr_flows--;
+ }
+
+ mutex_lock(&pfvf->mbox.lock);
+ req = otx2_mbox_alloc_msg_npc_mcam_free_entry(&pfvf->mbox);
+ if (!req) {
+ mutex_unlock(&pfvf->mbox.lock);
+ return -ENOMEM;
+ }
+
+ req->all = 1;
+ /* Send message to AF to free MCAM entries */
+ err = otx2_sync_mbox_msg(&pfvf->mbox);
+ if (err) {
+ mutex_unlock(&pfvf->mbox.lock);
+ return err;
+ }
+
+ pfvf->flags &= ~OTX2_FLAG_MCAM_ENTRIES_ALLOC;
+ mutex_unlock(&pfvf->mbox.lock);
+
+ return 0;
+}
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index 672768630557..9cf3e19cf1d8 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -1733,7 +1733,6 @@ static void otx2_do_set_rx_mode(struct work_struct *work)
req->mode = NIX_RX_MODE_UCAST;
- /* We don't support MAC address filtering yet */
if (netdev->flags & IFF_PROMISC)
req->mode |= NIX_RX_MODE_PROMISC;
else if (netdev->flags & (IFF_ALLMULTI | IFF_MULTICAST))
@@ -1747,11 +1746,16 @@ static int otx2_set_features(struct net_device *netdev,
netdev_features_t features)
{
netdev_features_t changed = features ^ netdev->features;
+ bool ntuple = !!(features & NETIF_F_NTUPLE);
struct otx2_nic *pf = netdev_priv(netdev);
if ((changed & NETIF_F_LOOPBACK) && netif_running(netdev))
return otx2_cgx_config_loopback(pf,
features & NETIF_F_LOOPBACK);
+
+ if ((changed & NETIF_F_NTUPLE) && !ntuple)
+ otx2_destroy_ntuple_flows(pf);
+
return 0;
}
@@ -2114,6 +2118,13 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
netdev->hw_features |= NETIF_F_LOOPBACK | NETIF_F_RXALL;
+ err = otx2_mcam_flow_init(pf);
+ if (err)
+ goto err_ptp_destroy;
+
+ if (pf->flags & OTX2_FLAG_NTUPLE_SUPPORT)
+ netdev->hw_features |= NETIF_F_NTUPLE;
+
netdev->gso_max_segs = OTX2_MAX_GSO_SEGS;
netdev->watchdog_timeo = OTX2_TX_TIMEOUT;
@@ -2126,7 +2137,7 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
err = register_netdev(netdev);
if (err) {
dev_err(dev, "Failed to register netdevice\n");
- goto err_ptp_destroy;
+ goto err_del_mcam_entries;
}
err = otx2_wq_init(pf);
@@ -2146,6 +2157,8 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
err_unreg_netdev:
unregister_netdev(netdev);
+err_del_mcam_entries:
+ otx2_mcam_flow_del(pf);
err_ptp_destroy:
otx2_ptp_destroy(pf);
err_detach_rsrc:
@@ -2304,6 +2317,7 @@ static void otx2_remove(struct pci_dev *pdev)
destroy_workqueue(pf->otx2_wq);
otx2_ptp_destroy(pf);
+ otx2_mcam_flow_del(pf);
otx2_detach_resources(&pf->mbox);
otx2_disable_mbox_intr(pf);
otx2_pfaf_mbox_destroy(pf);
--
2.16.5
^ permalink raw reply related [flat|nested] 24+ messages in thread* [PATCH v3 net-next 06/13] octeontx2-pf: Add support for unicast MAC address filtering
2020-11-11 7:13 [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Naveen Mamindlapalli
` (4 preceding siblings ...)
2020-11-11 7:13 ` [PATCH v3 net-next 05/13] octeontx2-pf: Add support for ethtool ntuple filters Naveen Mamindlapalli
@ 2020-11-11 7:13 ` Naveen Mamindlapalli
2020-11-11 7:13 ` [PATCH v3 net-next 07/13] octeontx2-af: Add debugfs entry to dump the MCAM rules Naveen Mamindlapalli
` (7 subsequent siblings)
13 siblings, 0 replies; 24+ messages in thread
From: Naveen Mamindlapalli @ 2020-11-11 7:13 UTC (permalink / raw)
To: netdev, linux-kernel
Cc: kuba, davem, saeed, sgoutham, lcherian, gakula, jerinj, sbhatta,
hkelam
From: Hariprasad Kelam <hkelam@marvell.com>
Add unicast MAC address filtering support using install flow
message. Total of 8 MCAM entries are allocated for adding
unicast mac filtering rules. If the MCAM allocation fails,
the unicast filtering support will not be advertised.
Signed-off-by: Hariprasad Kelam <hkelam@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
---
.../ethernet/marvell/octeontx2/nic/otx2_common.h | 15 ++-
.../ethernet/marvell/octeontx2/nic/otx2_flows.c | 120 ++++++++++++++++++++-
.../net/ethernet/marvell/octeontx2/nic/otx2_pf.c | 15 ++-
3 files changed, 143 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index 2387c40a2a8f..f959688e14a3 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -229,12 +229,21 @@ struct otx2_ptp {
#define OTX2_HW_TIMESTAMP_LEN 8
+struct otx2_mac_table {
+ u8 addr[ETH_ALEN];
+ u16 mcam_entry;
+ bool inuse;
+};
+
struct otx2_flow_config {
u16 entry[NPC_MAX_NONCONTIG_ENTRIES];
u32 nr_flows;
#define OTX2_MAX_NTUPLE_FLOWS 32
-#define OTX2_MCAM_COUNT OTX2_MAX_NTUPLE_FLOWS
+#define OTX2_MAX_UNICAST_FLOWS 8
+#define OTX2_MCAM_COUNT (OTX2_MAX_NTUPLE_FLOWS + \
+ OTX2_MAX_UNICAST_FLOWS)
u32 ntuple_offset;
+ u32 unicast_offset;
u32 ntuple_max_flows;
struct list_head flow_list;
};
@@ -251,6 +260,7 @@ struct otx2_nic {
#define OTX2_FLAG_INTF_DOWN BIT_ULL(2)
#define OTX2_FLAG_MCAM_ENTRIES_ALLOC BIT_ULL(3)
#define OTX2_FLAG_NTUPLE_SUPPORT BIT_ULL(4)
+#define OTX2_FLAG_UCAST_FLTR_SUPPORT BIT_ULL(5)
#define OTX2_FLAG_RX_PAUSE_ENABLED BIT_ULL(9)
#define OTX2_FLAG_TX_PAUSE_ENABLED BIT_ULL(10)
u64 flags;
@@ -279,6 +289,7 @@ struct otx2_nic {
struct refill_work *refill_wrk;
struct workqueue_struct *otx2_wq;
struct work_struct rx_mode_work;
+ struct otx2_mac_table *mac_table;
/* Ethtool stuff */
u32 msg_enable;
@@ -674,5 +685,7 @@ int otx2_add_flow(struct otx2_nic *pfvf,
int otx2_remove_flow(struct otx2_nic *pfvf, u32 location);
int otx2_prepare_flow_request(struct ethtool_rx_flow_spec *fsp,
struct npc_install_flow_req *req);
+int otx2_del_macfilter(struct net_device *netdev, const u8 *mac);
+int otx2_add_macfilter(struct net_device *netdev, const u8 *mac);
#endif /* OTX2_COMMON_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c
index 094f547c9889..7da80aa3f063 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c
@@ -46,14 +46,21 @@ int otx2_alloc_mcam_entries(struct otx2_nic *pfvf)
rsp = (struct npc_mcam_alloc_entry_rsp *)otx2_mbox_get_rsp
(&pfvf->mbox.mbox, 0, &req->hdr);
- if (rsp->count != req->count)
+ if (rsp->count != req->count) {
netdev_info(pfvf->netdev,
"Unable to allocate %d MCAM entries, got %d\n",
req->count, rsp->count);
-
- flow_cfg->ntuple_max_flows = rsp->count;
- flow_cfg->ntuple_offset = 0;
- pfvf->flags |= OTX2_FLAG_NTUPLE_SUPPORT;
+ /* support only ntuples here */
+ flow_cfg->ntuple_max_flows = rsp->count;
+ flow_cfg->ntuple_offset = 0;
+ pfvf->flags |= OTX2_FLAG_NTUPLE_SUPPORT;
+ } else {
+ flow_cfg->ntuple_offset = 0;
+ flow_cfg->unicast_offset = flow_cfg->ntuple_offset +
+ OTX2_MAX_NTUPLE_FLOWS;
+ pfvf->flags |= OTX2_FLAG_NTUPLE_SUPPORT;
+ pfvf->flags |= OTX2_FLAG_UCAST_FLTR_SUPPORT;
+ }
for (i = 0; i < rsp->count; i++)
flow_cfg->entry[i] = rsp->entry_list[i];
@@ -82,6 +89,11 @@ int otx2_mcam_flow_init(struct otx2_nic *pf)
if (err)
return err;
+ pf->mac_table = devm_kzalloc(pf->dev, sizeof(struct otx2_mac_table)
+ * OTX2_MAX_UNICAST_FLOWS, GFP_KERNEL);
+ if (!pf->mac_table)
+ return -ENOMEM;
+
return 0;
}
@@ -90,6 +102,104 @@ void otx2_mcam_flow_del(struct otx2_nic *pf)
otx2_destroy_mcam_flows(pf);
}
+/* On success adds mcam entry
+ * On failure enable promisous mode
+ */
+static int otx2_do_add_macfilter(struct otx2_nic *pf, const u8 *mac)
+{
+ struct otx2_flow_config *flow_cfg = pf->flow_cfg;
+ struct npc_install_flow_req *req;
+ int err, i;
+
+ if (!(pf->flags & OTX2_FLAG_UCAST_FLTR_SUPPORT))
+ return -ENOMEM;
+
+ /* dont have free mcam entries or uc list is greater than alloted */
+ if (netdev_uc_count(pf->netdev) > OTX2_MAX_UNICAST_FLOWS)
+ return -ENOMEM;
+
+ mutex_lock(&pf->mbox.lock);
+ req = otx2_mbox_alloc_msg_npc_install_flow(&pf->mbox);
+ if (!req) {
+ mutex_unlock(&pf->mbox.lock);
+ return -ENOMEM;
+ }
+
+ /* unicast offset starts with 32 0..31 for ntuple */
+ for (i = 0; i < OTX2_MAX_UNICAST_FLOWS; i++) {
+ if (pf->mac_table[i].inuse)
+ continue;
+ ether_addr_copy(pf->mac_table[i].addr, mac);
+ pf->mac_table[i].inuse = true;
+ pf->mac_table[i].mcam_entry =
+ flow_cfg->entry[i + flow_cfg->unicast_offset];
+ req->entry = pf->mac_table[i].mcam_entry;
+ break;
+ }
+
+ ether_addr_copy(req->packet.dmac, mac);
+ eth_broadcast_addr((u8 *)&req->mask.dmac);
+ req->features = BIT_ULL(NPC_DMAC);
+ req->channel = pf->hw.rx_chan_base;
+ req->intf = NIX_INTF_RX;
+ req->op = NIX_RX_ACTION_DEFAULT;
+ req->set_cntr = 1;
+
+ err = otx2_sync_mbox_msg(&pf->mbox);
+ mutex_unlock(&pf->mbox.lock);
+
+ return err;
+}
+
+int otx2_add_macfilter(struct net_device *netdev, const u8 *mac)
+{
+ struct otx2_nic *pf = netdev_priv(netdev);
+
+ return otx2_do_add_macfilter(pf, mac);
+}
+
+static bool otx2_get_mcamentry_for_mac(struct otx2_nic *pf, const u8 *mac,
+ int *mcam_entry)
+{
+ int i;
+
+ for (i = 0; i < OTX2_MAX_UNICAST_FLOWS; i++) {
+ if (!pf->mac_table[i].inuse)
+ continue;
+
+ if (ether_addr_equal(pf->mac_table[i].addr, mac)) {
+ *mcam_entry = pf->mac_table[i].mcam_entry;
+ pf->mac_table[i].inuse = false;
+ return true;
+ }
+ }
+ return false;
+}
+
+int otx2_del_macfilter(struct net_device *netdev, const u8 *mac)
+{
+ struct otx2_nic *pf = netdev_priv(netdev);
+ struct npc_delete_flow_req *req;
+ int err, mcam_entry;
+
+ /* check does mcam entry exists for given mac */
+ if (!otx2_get_mcamentry_for_mac(pf, mac, &mcam_entry))
+ return 0;
+
+ mutex_lock(&pf->mbox.lock);
+ req = otx2_mbox_alloc_msg_npc_delete_flow(&pf->mbox);
+ if (!req) {
+ mutex_unlock(&pf->mbox.lock);
+ return -ENOMEM;
+ }
+ req->entry = mcam_entry;
+ /* Send message to AF */
+ err = otx2_sync_mbox_msg(&pf->mbox);
+ mutex_unlock(&pf->mbox.lock);
+
+ return err;
+}
+
static struct otx2_flow *otx2_find_flow(struct otx2_nic *pfvf, u32 location)
{
struct otx2_flow *iter;
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index 9cf3e19cf1d8..ac90302b1d81 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -1720,10 +1720,20 @@ static void otx2_do_set_rx_mode(struct work_struct *work)
struct otx2_nic *pf = container_of(work, struct otx2_nic, rx_mode_work);
struct net_device *netdev = pf->netdev;
struct nix_rx_mode *req;
+ bool promisc = false;
if (!(netdev->flags & IFF_UP))
return;
+ if ((netdev->flags & IFF_PROMISC) ||
+ (netdev_uc_count(netdev) > OTX2_MAX_UNICAST_FLOWS)) {
+ promisc = true;
+ }
+
+ /* Write unicast address to mcam entries or del from mcam */
+ if (!promisc && netdev->priv_flags & IFF_UNICAST_FLT)
+ __dev_uc_sync(netdev, otx2_add_macfilter, otx2_del_macfilter);
+
mutex_lock(&pf->mbox.lock);
req = otx2_mbox_alloc_msg_nix_set_rx_mode(&pf->mbox);
if (!req) {
@@ -1733,7 +1743,7 @@ static void otx2_do_set_rx_mode(struct work_struct *work)
req->mode = NIX_RX_MODE_UCAST;
- if (netdev->flags & IFF_PROMISC)
+ if (promisc)
req->mode |= NIX_RX_MODE_PROMISC;
else if (netdev->flags & (IFF_ALLMULTI | IFF_MULTICAST))
req->mode |= NIX_RX_MODE_ALLMULTI;
@@ -2125,6 +2135,9 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
if (pf->flags & OTX2_FLAG_NTUPLE_SUPPORT)
netdev->hw_features |= NETIF_F_NTUPLE;
+ if (pf->flags & OTX2_FLAG_UCAST_FLTR_SUPPORT)
+ netdev->priv_flags |= IFF_UNICAST_FLT;
+
netdev->gso_max_segs = OTX2_MAX_GSO_SEGS;
netdev->watchdog_timeo = OTX2_TX_TIMEOUT;
--
2.16.5
^ permalink raw reply related [flat|nested] 24+ messages in thread* [PATCH v3 net-next 07/13] octeontx2-af: Add debugfs entry to dump the MCAM rules
2020-11-11 7:13 [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Naveen Mamindlapalli
` (5 preceding siblings ...)
2020-11-11 7:13 ` [PATCH v3 net-next 06/13] octeontx2-pf: Add support for unicast MAC address filtering Naveen Mamindlapalli
@ 2020-11-11 7:13 ` Naveen Mamindlapalli
2020-11-12 19:48 ` Saeed Mahameed
2020-11-11 7:13 ` [PATCH v3 net-next 08/13] octeontx2-af: Modify nix_vtag_cfg mailbox to support TX VTAG entries Naveen Mamindlapalli
` (6 subsequent siblings)
13 siblings, 1 reply; 24+ messages in thread
From: Naveen Mamindlapalli @ 2020-11-11 7:13 UTC (permalink / raw)
To: netdev, linux-kernel
Cc: kuba, davem, saeed, sgoutham, lcherian, gakula, jerinj, sbhatta,
hkelam
From: Subbaraya Sundeep <sbhatta@marvell.com>
Add debugfs support to dump the MCAM rules installed using
NPC_INSTALL_FLOW mbox message. Debugfs file can display mcam
entry, counter if any, flow type and counter hits.
Ethtool will dump the ntuple flows related to the PF only.
The debugfs file gives systemwide view of the MCAM rules
installed by all the PF's.
Below is the example output when the debugfs file is read:
~ # mount -t debugfs none /sys/kernel/debug
~ # cat /sys/kernel/debug/octeontx2/npc/mcam_rules
Installed by: PF1
direction: RX
mcam entry: 227
udp source port 23 mask 0xffff
Forward to: PF1 VF0
action: Direct to queue 0
enabled: yes
counter: 1
hits: 0
Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
---
.../ethernet/marvell/octeontx2/af/rvu_debugfs.c | 197 +++++++++++++++++++++
1 file changed, 197 insertions(+)
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c
index b7b6b6f8865a..39e1a614aaf8 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_debugfs.c
@@ -1770,6 +1770,198 @@ static int rvu_dbg_npc_rx_miss_stats_display(struct seq_file *filp,
RVU_DEBUG_SEQ_FOPS(npc_rx_miss_act, npc_rx_miss_stats_display, NULL);
+static void rvu_dbg_npc_mcam_show_flows(struct seq_file *s,
+ struct rvu_npc_mcam_rule *rule)
+{
+ u8 bit;
+
+ for_each_set_bit(bit, (unsigned long *)&rule->features, 64) {
+ seq_printf(s, "\t%s ", npc_get_field_name(bit));
+ switch (bit) {
+ case NPC_DMAC:
+ seq_printf(s, "%pM ", rule->packet.dmac);
+ seq_printf(s, "mask %pM\n", rule->mask.dmac);
+ break;
+ case NPC_SMAC:
+ seq_printf(s, "%pM ", rule->packet.smac);
+ seq_printf(s, "mask %pM\n", rule->mask.smac);
+ break;
+ case NPC_ETYPE:
+ seq_printf(s, "0x%x ", ntohs(rule->packet.etype));
+ seq_printf(s, "mask 0x%x\n", ntohs(rule->mask.etype));
+ break;
+ case NPC_OUTER_VID:
+ seq_printf(s, "%d ", ntohs(rule->packet.vlan_tci));
+ seq_printf(s, "mask 0x%x\n",
+ ntohs(rule->mask.vlan_tci));
+ break;
+ case NPC_TOS:
+ seq_printf(s, "%d ", rule->packet.tos);
+ seq_printf(s, "mask 0x%x\n", rule->mask.tos);
+ break;
+ case NPC_SIP_IPV4:
+ seq_printf(s, "%pI4 ", &rule->packet.ip4src);
+ seq_printf(s, "mask %pI4\n", &rule->mask.ip4src);
+ break;
+ case NPC_DIP_IPV4:
+ seq_printf(s, "%pI4 ", &rule->packet.ip4dst);
+ seq_printf(s, "mask %pI4\n", &rule->mask.ip4dst);
+ break;
+ case NPC_SIP_IPV6:
+ seq_printf(s, "%pI6 ", rule->packet.ip6src);
+ seq_printf(s, "mask %pI6\n", rule->mask.ip6src);
+ break;
+ case NPC_DIP_IPV6:
+ seq_printf(s, "%pI6 ", rule->packet.ip6dst);
+ seq_printf(s, "mask %pI6\n", rule->mask.ip6dst);
+ break;
+ case NPC_SPORT_TCP:
+ case NPC_SPORT_UDP:
+ case NPC_SPORT_SCTP:
+ seq_printf(s, "%d ", ntohs(rule->packet.sport));
+ seq_printf(s, "mask 0x%x\n", ntohs(rule->mask.sport));
+ break;
+ case NPC_DPORT_TCP:
+ case NPC_DPORT_UDP:
+ case NPC_DPORT_SCTP:
+ seq_printf(s, "%d ", ntohs(rule->packet.dport));
+ seq_printf(s, "mask 0x%x\n", ntohs(rule->mask.dport));
+ break;
+ default:
+ break;
+ }
+ }
+}
+
+static void rvu_dbg_npc_mcam_show_action(struct seq_file *s,
+ struct rvu_npc_mcam_rule *rule)
+{
+ if (rule->intf == NIX_INTF_TX) {
+ switch (rule->tx_action.op) {
+ case NIX_TX_ACTIONOP_DROP:
+ seq_puts(s, "\taction: Drop\n");
+ break;
+ case NIX_TX_ACTIONOP_UCAST_DEFAULT:
+ seq_puts(s, "\taction: Unicast to default channel\n");
+ break;
+ case NIX_TX_ACTIONOP_UCAST_CHAN:
+ seq_printf(s, "\taction: Unicast to channel %d\n",
+ rule->tx_action.index);
+ break;
+ case NIX_TX_ACTIONOP_MCAST:
+ seq_puts(s, "\taction: Multicast\n");
+ break;
+ case NIX_TX_ACTIONOP_DROP_VIOL:
+ seq_puts(s, "\taction: Lockdown Violation Drop\n");
+ break;
+ default:
+ break;
+ };
+ } else {
+ switch (rule->rx_action.op) {
+ case NIX_RX_ACTIONOP_DROP:
+ seq_puts(s, "\taction: Drop\n");
+ break;
+ case NIX_RX_ACTIONOP_UCAST:
+ seq_printf(s, "\taction: Direct to queue %d\n",
+ rule->rx_action.index);
+ break;
+ case NIX_RX_ACTIONOP_RSS:
+ seq_puts(s, "\taction: RSS\n");
+ break;
+ case NIX_RX_ACTIONOP_UCAST_IPSEC:
+ seq_puts(s, "\taction: Unicast ipsec\n");
+ break;
+ case NIX_RX_ACTIONOP_MCAST:
+ seq_puts(s, "\taction: Multicast\n");
+ break;
+ default:
+ break;
+ };
+ }
+}
+
+static const char *rvu_dbg_get_intf_name(int intf)
+{
+ switch (intf) {
+ case NIX_INTFX_RX(0):
+ return "NIX0_RX";
+ case NIX_INTFX_RX(1):
+ return "NIX1_RX";
+ case NIX_INTFX_TX(0):
+ return "NIX0_TX";
+ case NIX_INTFX_TX(1):
+ return "NIX1_TX";
+ default:
+ break;
+ }
+
+ return "unknown";
+}
+
+static int rvu_dbg_npc_mcam_show_rules(struct seq_file *s, void *unused)
+{
+ struct rvu_npc_mcam_rule *iter;
+ struct rvu *rvu = s->private;
+ struct npc_mcam *mcam;
+ int pf, vf = -1;
+ int blkaddr;
+ u16 target;
+ u64 hits;
+
+ blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NPC, 0);
+ if (blkaddr < 0)
+ return 0;
+
+ mcam = &rvu->hw->mcam;
+
+ mutex_lock(&mcam->lock);
+ list_for_each_entry(iter, &mcam->mcam_rules, list) {
+ pf = (iter->owner >> RVU_PFVF_PF_SHIFT) & RVU_PFVF_PF_MASK;
+ seq_printf(s, "\n\tInstalled by: PF%d ", pf);
+
+ if (iter->owner & RVU_PFVF_FUNC_MASK) {
+ vf = (iter->owner & RVU_PFVF_FUNC_MASK) - 1;
+ seq_printf(s, "VF%d", vf);
+ }
+ seq_puts(s, "\n");
+
+ seq_printf(s, "\tdirection: %s\n", is_npc_intf_rx(iter->intf) ?
+ "RX" : "TX");
+ seq_printf(s, "\tinterface: %s\n",
+ rvu_dbg_get_intf_name(iter->intf));
+ seq_printf(s, "\tmcam entry: %d\n", iter->entry);
+
+ rvu_dbg_npc_mcam_show_flows(s, iter);
+ if (iter->intf == NIX_INTF_RX) {
+ target = iter->rx_action.pf_func;
+ pf = (target >> RVU_PFVF_PF_SHIFT) & RVU_PFVF_PF_MASK;
+ seq_printf(s, "\tForward to: PF%d ", pf);
+
+ if (target & RVU_PFVF_FUNC_MASK) {
+ vf = (target & RVU_PFVF_FUNC_MASK) - 1;
+ seq_printf(s, "VF%d", vf);
+ }
+ seq_puts(s, "\n");
+ }
+
+ rvu_dbg_npc_mcam_show_action(s, iter);
+ seq_printf(s, "\tenabled: %s\n", iter->enable ? "yes" : "no");
+
+ if (!iter->has_cntr)
+ continue;
+ seq_printf(s, "\tcounter: %d\n", iter->cntr);
+
+ hits = rvu_read64(rvu, blkaddr, NPC_AF_MATCH_STATX(iter->cntr));
+ seq_printf(s, "\thits: %lld\n", hits);
+ }
+ mutex_unlock(&mcam->lock);
+
+ return 0;
+}
+
+RVU_DEBUG_SEQ_FOPS(npc_mcam_rules, npc_mcam_show_rules, NULL);
+
static void rvu_dbg_npc_init(struct rvu *rvu)
{
const struct device *dev = &rvu->pdev->dev;
@@ -1784,6 +1976,11 @@ static void rvu_dbg_npc_init(struct rvu *rvu)
if (!pfile)
goto create_failed;
+ pfile = debugfs_create_file("mcam_rules", 0444, rvu->rvu_dbg.npc,
+ rvu, &rvu_dbg_npc_mcam_rules_fops);
+ if (!pfile)
+ goto create_failed;
+
pfile = debugfs_create_file("rx_miss_act_stats", 0444, rvu->rvu_dbg.npc,
rvu, &rvu_dbg_npc_rx_miss_act_fops);
if (!pfile)
--
2.16.5
^ permalink raw reply related [flat|nested] 24+ messages in thread* Re: [PATCH v3 net-next 07/13] octeontx2-af: Add debugfs entry to dump the MCAM rules
2020-11-11 7:13 ` [PATCH v3 net-next 07/13] octeontx2-af: Add debugfs entry to dump the MCAM rules Naveen Mamindlapalli
@ 2020-11-12 19:48 ` Saeed Mahameed
0 siblings, 0 replies; 24+ messages in thread
From: Saeed Mahameed @ 2020-11-12 19:48 UTC (permalink / raw)
To: Naveen Mamindlapalli, netdev, linux-kernel
Cc: kuba, davem, sgoutham, lcherian, gakula, jerinj, sbhatta, hkelam
On Wed, 2020-11-11 at 12:43 +0530, Naveen Mamindlapalli wrote:
> From: Subbaraya Sundeep <sbhatta@marvell.com>
>
> Add debugfs support to dump the MCAM rules installed using
> NPC_INSTALL_FLOW mbox message. Debugfs file can display mcam
> entry, counter if any, flow type and counter hits.
>
> Ethtool will dump the ntuple flows related to the PF only.
> The debugfs file gives systemwide view of the MCAM rules
> installed by all the PF's.
>
> Below is the example output when the debugfs file is read:
> ~ # mount -t debugfs none /sys/kernel/debug
> ~ # cat /sys/kernel/debug/octeontx2/npc/mcam_rules
>
> Installed by: PF1
> direction: RX
> mcam entry: 227
> udp source port 23 mask 0xffff
> Forward to: PF1 VF0
> action: Direct to queue 0
> enabled: yes
> counter: 1
> hits: 0
>
I don't want to block this series or anything, but you might want to
use devlink dpipe interface for this:
https://www.kernel.org/doc/html/latest/networking/devlink/devlink-dpipe.html
As a future patch of course.
Thanks,
Saeed.
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH v3 net-next 08/13] octeontx2-af: Modify nix_vtag_cfg mailbox to support TX VTAG entries
2020-11-11 7:13 [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Naveen Mamindlapalli
` (6 preceding siblings ...)
2020-11-11 7:13 ` [PATCH v3 net-next 07/13] octeontx2-af: Add debugfs entry to dump the MCAM rules Naveen Mamindlapalli
@ 2020-11-11 7:13 ` Naveen Mamindlapalli
2020-11-11 7:14 ` [PATCH v3 net-next 09/13] octeontx2-pf: Implement ingress/egress VLAN offload Naveen Mamindlapalli
` (5 subsequent siblings)
13 siblings, 0 replies; 24+ messages in thread
From: Naveen Mamindlapalli @ 2020-11-11 7:13 UTC (permalink / raw)
To: netdev, linux-kernel
Cc: kuba, davem, saeed, sgoutham, lcherian, gakula, jerinj, sbhatta,
hkelam
From: Vamsi Attunuru <vattunuru@marvell.com>
This patch modifies the existing nix_vtag_config mailbox message
to allocate and free TX VTAG entries as requested by a NIX PF.
The TX VTAG entries are global resource that shared by all PFs
and each entry specifies the size of VTAG to insert and the VTAG
header data to insert. The mailbox response contains the entry
index which is used by mailbox requester in configuring the
NPC_TX_VTAG_ACTION for any MCAM entry.
Signed-off-by: Vamsi Attunuru <vattunuru@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
---
drivers/net/ethernet/marvell/octeontx2/af/mbox.h | 60 +++++--
drivers/net/ethernet/marvell/octeontx2/af/rvu.h | 8 +
.../net/ethernet/marvell/octeontx2/af/rvu_nix.c | 195 ++++++++++++++++++++-
3 files changed, 250 insertions(+), 13 deletions(-)
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
index ac3118d2d126..4f230a7272ce 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
@@ -204,7 +204,8 @@ M(NIX_TXSCH_ALLOC, 0x8004, nix_txsch_alloc, \
M(NIX_TXSCH_FREE, 0x8005, nix_txsch_free, nix_txsch_free_req, msg_rsp) \
M(NIX_TXSCHQ_CFG, 0x8006, nix_txschq_cfg, nix_txschq_config, msg_rsp) \
M(NIX_STATS_RST, 0x8007, nix_stats_rst, msg_req, msg_rsp) \
-M(NIX_VTAG_CFG, 0x8008, nix_vtag_cfg, nix_vtag_config, msg_rsp) \
+M(NIX_VTAG_CFG, 0x8008, nix_vtag_cfg, nix_vtag_config, \
+ nix_vtag_config_rsp) \
M(NIX_RSS_FLOWKEY_CFG, 0x8009, nix_rss_flowkey_cfg, \
nix_rss_flowkey_cfg, \
nix_rss_flowkey_cfg_rsp) \
@@ -477,6 +478,7 @@ enum nix_af_status {
NIX_AF_ERR_LSO_CFG_FAIL = -418,
NIX_AF_INVAL_NPA_PF_FUNC = -419,
NIX_AF_INVAL_SSO_PF_FUNC = -420,
+ NIX_AF_ERR_TX_VTAG_NOSPC = -421,
};
/* For NIX LF context alloc and init */
@@ -516,7 +518,8 @@ struct nix_lf_alloc_rsp {
struct nix_lf_free_req {
struct mbox_msghdr hdr;
-#define NIX_LF_DISABLE_FLOWS BIT_ULL(0)
+#define NIX_LF_DISABLE_FLOWS BIT_ULL(0)
+#define NIX_LF_DONT_FREE_TX_VTAG BIT_ULL(1)
u64 flags;
};
@@ -610,14 +613,40 @@ struct nix_vtag_config {
union {
/* valid when cfg_type is '0' */
struct {
- /* tx vlan0 tag(C-VLAN) */
- u64 vlan0;
- /* tx vlan1 tag(S-VLAN) */
- u64 vlan1;
- /* insert tx vlan tag */
- u8 insert_vlan :1;
- /* insert tx double vlan tag */
- u8 double_vlan :1;
+ u64 vtag0;
+ u64 vtag1;
+
+ /* cfg_vtag0 & cfg_vtag1 fields are valid
+ * when free_vtag0 & free_vtag1 are '0's.
+ */
+ /* cfg_vtag0 = 1 to configure vtag0 */
+ u8 cfg_vtag0 :1;
+ /* cfg_vtag1 = 1 to configure vtag1 */
+ u8 cfg_vtag1 :1;
+
+ /* vtag0_idx & vtag1_idx are only valid when
+ * both cfg_vtag0 & cfg_vtag1 are '0's,
+ * these fields are used along with free_vtag0
+ * & free_vtag1 to free the nix lf's tx_vlan
+ * configuration.
+ *
+ * Denotes the indices of tx_vtag def registers
+ * that needs to be cleared and freed.
+ */
+ int vtag0_idx;
+ int vtag1_idx;
+
+ /* free_vtag0 & free_vtag1 fields are valid
+ * when cfg_vtag0 & cfg_vtag1 are '0's.
+ */
+ /* free_vtag0 = 1 clears vtag0 configuration
+ * vtag0_idx denotes the index to be cleared.
+ */
+ u8 free_vtag0 :1;
+ /* free_vtag1 = 1 clears vtag1 configuration
+ * vtag1_idx denotes the index to be cleared.
+ */
+ u8 free_vtag1 :1;
} tx;
/* valid when cfg_type is '1' */
@@ -632,6 +661,17 @@ struct nix_vtag_config {
};
};
+struct nix_vtag_config_rsp {
+ struct mbox_msghdr hdr;
+ int vtag0_idx;
+ int vtag1_idx;
+ /* Indices of tx_vtag def registers used to configure
+ * tx vtag0 & vtag1 headers, these indices are valid
+ * when nix_vtag_config mbox requested for vtag0 and/
+ * or vtag1 configuration.
+ */
+};
+
struct nix_rss_flowkey_cfg {
struct mbox_msghdr hdr;
int mcam_index; /* MCAM entry index to modify */
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
index fecde9e73a33..6a45573724b4 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
@@ -264,6 +264,13 @@ struct nix_lso {
u8 in_use;
};
+struct nix_txvlan {
+#define NIX_TX_VTAG_DEF_MAX 0x400
+ struct rsrc_bmap rsrc;
+ u16 *entry2pfvf_map;
+ struct mutex rsrc_lock; /* Serialize resource alloc/free */
+};
+
struct nix_hw {
int blkaddr;
struct rvu *rvu;
@@ -272,6 +279,7 @@ struct nix_hw {
struct nix_flowkey flowkey;
struct nix_mark_format mark_format;
struct nix_lso lso;
+ struct nix_txvlan txvlan;
};
/* RVU block's capabilities or functionality,
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
index 9314309da539..4709d8b6197b 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
@@ -17,6 +17,7 @@
#include "npc.h"
#include "cgx.h"
+static void nix_free_tx_vtag_entries(struct rvu *rvu, u16 pcifunc);
static int rvu_nix_get_bpid(struct rvu *rvu, struct nix_bp_cfg_req *req,
int type, int chan_id);
@@ -1253,6 +1254,10 @@ int rvu_mbox_handler_nix_lf_free(struct rvu *rvu, struct nix_lf_free_req *req,
else
rvu_npc_free_mcam_entries(rvu, pcifunc, nixlf);
+ /* Free any tx vtag def entries used by this NIX LF */
+ if (!(req->flags & NIX_LF_DONT_FREE_TX_VTAG))
+ nix_free_tx_vtag_entries(rvu, pcifunc);
+
nix_interface_deinit(rvu, pcifunc, nixlf);
/* Reset this NIX LF */
@@ -1994,9 +1999,149 @@ static int nix_rx_vtag_cfg(struct rvu *rvu, int nixlf, int blkaddr,
return 0;
}
+static int nix_tx_vtag_free(struct rvu *rvu, int blkaddr,
+ u16 pcifunc, int index)
+{
+ struct nix_hw *nix_hw = get_nix_hw(rvu->hw, blkaddr);
+ struct nix_txvlan *vlan = &nix_hw->txvlan;
+
+ if (vlan->entry2pfvf_map[index] != pcifunc)
+ return NIX_AF_ERR_PARAM;
+
+ rvu_write64(rvu, blkaddr,
+ NIX_AF_TX_VTAG_DEFX_DATA(index), 0x0ull);
+ rvu_write64(rvu, blkaddr,
+ NIX_AF_TX_VTAG_DEFX_CTL(index), 0x0ull);
+
+ vlan->entry2pfvf_map[index] = 0;
+ rvu_free_rsrc(&vlan->rsrc, index);
+
+ return 0;
+}
+
+static void nix_free_tx_vtag_entries(struct rvu *rvu, u16 pcifunc)
+{
+ struct nix_txvlan *vlan;
+ struct nix_hw *nix_hw;
+ int index, blkaddr;
+
+ blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NIX, pcifunc);
+ if (blkaddr < 0)
+ return;
+
+ nix_hw = get_nix_hw(rvu->hw, blkaddr);
+ vlan = &nix_hw->txvlan;
+
+ mutex_lock(&vlan->rsrc_lock);
+ /* Scan all the entries and free the ones mapped to 'pcifunc' */
+ for (index = 0; index < vlan->rsrc.max; index++) {
+ if (vlan->entry2pfvf_map[index] == pcifunc)
+ nix_tx_vtag_free(rvu, blkaddr, pcifunc, index);
+ }
+ mutex_unlock(&vlan->rsrc_lock);
+}
+
+static int nix_tx_vtag_alloc(struct rvu *rvu, int blkaddr,
+ u64 vtag, u8 size)
+{
+ struct nix_hw *nix_hw = get_nix_hw(rvu->hw, blkaddr);
+ struct nix_txvlan *vlan = &nix_hw->txvlan;
+ u64 regval;
+ int index;
+
+ mutex_lock(&vlan->rsrc_lock);
+
+ index = rvu_alloc_rsrc(&vlan->rsrc);
+ if (index < 0) {
+ mutex_unlock(&vlan->rsrc_lock);
+ return index;
+ }
+
+ mutex_unlock(&vlan->rsrc_lock);
+
+ regval = size ? vtag : vtag << 32;
+
+ rvu_write64(rvu, blkaddr,
+ NIX_AF_TX_VTAG_DEFX_DATA(index), regval);
+ rvu_write64(rvu, blkaddr,
+ NIX_AF_TX_VTAG_DEFX_CTL(index), size);
+
+ return index;
+}
+
+static int nix_tx_vtag_decfg(struct rvu *rvu, int blkaddr,
+ struct nix_vtag_config *req)
+{
+ struct nix_hw *nix_hw = get_nix_hw(rvu->hw, blkaddr);
+ struct nix_txvlan *vlan = &nix_hw->txvlan;
+ u16 pcifunc = req->hdr.pcifunc;
+ int idx0 = req->tx.vtag0_idx;
+ int idx1 = req->tx.vtag1_idx;
+ int err;
+
+ if (req->tx.free_vtag0 && req->tx.free_vtag1)
+ if (vlan->entry2pfvf_map[idx0] != pcifunc ||
+ vlan->entry2pfvf_map[idx1] != pcifunc)
+ return NIX_AF_ERR_PARAM;
+
+ mutex_lock(&vlan->rsrc_lock);
+
+ if (req->tx.free_vtag0) {
+ err = nix_tx_vtag_free(rvu, blkaddr, pcifunc, idx0);
+ if (err)
+ goto exit;
+ }
+
+ if (req->tx.free_vtag1)
+ err = nix_tx_vtag_free(rvu, blkaddr, pcifunc, idx1);
+
+exit:
+ mutex_unlock(&vlan->rsrc_lock);
+ return err;
+}
+
+static int nix_tx_vtag_cfg(struct rvu *rvu, int blkaddr,
+ struct nix_vtag_config *req,
+ struct nix_vtag_config_rsp *rsp)
+{
+ struct nix_hw *nix_hw = get_nix_hw(rvu->hw, blkaddr);
+ struct nix_txvlan *vlan = &nix_hw->txvlan;
+ u16 pcifunc = req->hdr.pcifunc;
+
+ if (req->tx.cfg_vtag0) {
+ rsp->vtag0_idx =
+ nix_tx_vtag_alloc(rvu, blkaddr,
+ req->tx.vtag0, req->vtag_size);
+
+ if (rsp->vtag0_idx < 0)
+ return NIX_AF_ERR_TX_VTAG_NOSPC;
+
+ vlan->entry2pfvf_map[rsp->vtag0_idx] = pcifunc;
+ }
+
+ if (req->tx.cfg_vtag1) {
+ rsp->vtag1_idx =
+ nix_tx_vtag_alloc(rvu, blkaddr,
+ req->tx.vtag1, req->vtag_size);
+
+ if (rsp->vtag1_idx < 0)
+ goto err_free;
+
+ vlan->entry2pfvf_map[rsp->vtag1_idx] = pcifunc;
+ }
+
+ return 0;
+
+err_free:
+ if (req->tx.cfg_vtag0)
+ nix_tx_vtag_free(rvu, blkaddr, pcifunc, rsp->vtag0_idx);
+
+ return NIX_AF_ERR_TX_VTAG_NOSPC;
+}
+
int rvu_mbox_handler_nix_vtag_cfg(struct rvu *rvu,
struct nix_vtag_config *req,
- struct msg_rsp *rsp)
+ struct nix_vtag_config_rsp *rsp)
{
u16 pcifunc = req->hdr.pcifunc;
int blkaddr, nixlf, err;
@@ -2006,12 +2151,21 @@ int rvu_mbox_handler_nix_vtag_cfg(struct rvu *rvu,
return err;
if (req->cfg_type) {
+ /* rx vtag configuration */
err = nix_rx_vtag_cfg(rvu, nixlf, blkaddr, req);
if (err)
return NIX_AF_ERR_PARAM;
} else {
- /* TODO: handle tx vtag configuration */
- return 0;
+ /* tx vtag configuration */
+ if ((req->tx.cfg_vtag0 || req->tx.cfg_vtag1) &&
+ (req->tx.free_vtag0 || req->tx.free_vtag1))
+ return NIX_AF_ERR_PARAM;
+
+ if (req->tx.cfg_vtag0 || req->tx.cfg_vtag1)
+ return nix_tx_vtag_cfg(rvu, blkaddr, req, rsp);
+
+ if (req->tx.free_vtag0 || req->tx.free_vtag1)
+ return nix_tx_vtag_decfg(rvu, blkaddr, req);
}
return 0;
@@ -2249,6 +2403,31 @@ static int nix_setup_mcast(struct rvu *rvu, struct nix_hw *nix_hw, int blkaddr)
return nix_setup_bcast_tables(rvu, nix_hw);
}
+static int nix_setup_txvlan(struct rvu *rvu, struct nix_hw *nix_hw)
+{
+ struct nix_txvlan *vlan = &nix_hw->txvlan;
+ int err;
+
+ /* Allocate resource bimap for tx vtag def registers*/
+ vlan->rsrc.max = NIX_TX_VTAG_DEF_MAX;
+ err = rvu_alloc_bitmap(&vlan->rsrc);
+ if (err)
+ return -ENOMEM;
+
+ /* Alloc memory for saving entry to RVU PFFUNC allocation mapping */
+ vlan->entry2pfvf_map = devm_kcalloc(rvu->dev, vlan->rsrc.max,
+ sizeof(u16), GFP_KERNEL);
+ if (!vlan->entry2pfvf_map)
+ goto free_mem;
+
+ mutex_init(&vlan->rsrc_lock);
+ return 0;
+
+free_mem:
+ kfree(vlan->rsrc.bmap);
+ return -ENOMEM;
+}
+
static int nix_setup_txschq(struct rvu *rvu, struct nix_hw *nix_hw, int blkaddr)
{
struct nix_txsch *txsch;
@@ -3243,6 +3422,10 @@ static int rvu_nix_block_init(struct rvu *rvu, struct nix_hw *nix_hw)
if (err)
return err;
+ err = nix_setup_txvlan(rvu, nix_hw);
+ if (err)
+ return err;
+
/* Configure segmentation offload formats */
nix_setup_lso(rvu, nix_hw, blkaddr);
@@ -3329,6 +3512,7 @@ static void rvu_nix_block_freemem(struct rvu *rvu, int blkaddr,
{
struct nix_txsch *txsch;
struct nix_mcast *mcast;
+ struct nix_txvlan *vlan;
struct nix_hw *nix_hw;
int lvl;
@@ -3344,6 +3528,11 @@ static void rvu_nix_block_freemem(struct rvu *rvu, int blkaddr,
kfree(txsch->schq.bmap);
}
+ vlan = &nix_hw->txvlan;
+ kfree(vlan->rsrc.bmap);
+ mutex_destroy(&vlan->rsrc_lock);
+ devm_kfree(rvu->dev, vlan->entry2pfvf_map);
+
mcast = &nix_hw->mcast;
qmem_free(rvu->dev, mcast->mce_ctx);
qmem_free(rvu->dev, mcast->mcast_buf);
--
2.16.5
^ permalink raw reply related [flat|nested] 24+ messages in thread* [PATCH v3 net-next 09/13] octeontx2-pf: Implement ingress/egress VLAN offload
2020-11-11 7:13 [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Naveen Mamindlapalli
` (7 preceding siblings ...)
2020-11-11 7:13 ` [PATCH v3 net-next 08/13] octeontx2-af: Modify nix_vtag_cfg mailbox to support TX VTAG entries Naveen Mamindlapalli
@ 2020-11-11 7:14 ` Naveen Mamindlapalli
2020-11-11 7:14 ` [PATCH v3 net-next 10/13] octeontx2-pf: Add support for SR-IOV management functions Naveen Mamindlapalli
` (4 subsequent siblings)
13 siblings, 0 replies; 24+ messages in thread
From: Naveen Mamindlapalli @ 2020-11-11 7:14 UTC (permalink / raw)
To: netdev, linux-kernel
Cc: kuba, davem, saeed, sgoutham, lcherian, gakula, jerinj, sbhatta,
hkelam
From: Hariprasad Kelam <hkelam@marvell.com>
This patch implements egress VLAN offload by appending NIX_SEND_EXT_S
header to NIX_SEND_HDR_S. The VLAN TCI information is specified
in the NIX_SEND_EXT_S. The VLAN offload in the ingress path is
implemented by configuring the NIX_RX_VTAG_ACTION_S to strip and
capture the outer vlan fields. The NIX PF allocates one MCAM entry
for Rx VLAN offload.
Signed-off-by: Hariprasad Kelam <hkelam@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
---
drivers/net/ethernet/marvell/octeontx2/af/mbox.h | 13 +++
.../net/ethernet/marvell/octeontx2/af/rvu_nix.c | 3 +-
.../ethernet/marvell/octeontx2/nic/otx2_common.c | 8 +-
.../ethernet/marvell/octeontx2/nic/otx2_common.h | 8 +-
.../ethernet/marvell/octeontx2/nic/otx2_flows.c | 102 +++++++++++++++++++++
.../net/ethernet/marvell/octeontx2/nic/otx2_pf.c | 16 ++++
.../net/ethernet/marvell/octeontx2/nic/otx2_txrx.c | 16 ++++
.../net/ethernet/marvell/octeontx2/nic/otx2_vf.c | 5 +
8 files changed, 167 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
index 4f230a7272ce..ef20078508a5 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
@@ -479,6 +479,19 @@ enum nix_af_status {
NIX_AF_INVAL_NPA_PF_FUNC = -419,
NIX_AF_INVAL_SSO_PF_FUNC = -420,
NIX_AF_ERR_TX_VTAG_NOSPC = -421,
+ NIX_AF_ERR_RX_VTAG_INUSE = -422,
+};
+
+/* For NIX RX vtag action */
+enum nix_rx_vtag0_type {
+ NIX_AF_LFX_RX_VTAG_TYPE0, /* reserved for rx vlan offload */
+ NIX_AF_LFX_RX_VTAG_TYPE1,
+ NIX_AF_LFX_RX_VTAG_TYPE2,
+ NIX_AF_LFX_RX_VTAG_TYPE3,
+ NIX_AF_LFX_RX_VTAG_TYPE4,
+ NIX_AF_LFX_RX_VTAG_TYPE5,
+ NIX_AF_LFX_RX_VTAG_TYPE6,
+ NIX_AF_LFX_RX_VTAG_TYPE7,
};
/* For NIX LF context alloc and init */
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
index 4709d8b6197b..97a8f932d1e2 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
@@ -1986,7 +1986,8 @@ static int nix_rx_vtag_cfg(struct rvu *rvu, int nixlf, int blkaddr,
{
u64 regval = req->vtag_size;
- if (req->rx.vtag_type > 7 || req->vtag_size > VTAGSIZE_T8)
+ if (req->rx.vtag_type > NIX_AF_LFX_RX_VTAG_TYPE7 ||
+ req->vtag_size > VTAGSIZE_T8)
return -EINVAL;
if (req->rx.capture_vtag)
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
index 9f3d6715748e..68fb4e4757aa 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c
@@ -191,10 +191,14 @@ int otx2_set_mac_address(struct net_device *netdev, void *p)
if (!is_valid_ether_addr(addr->sa_data))
return -EADDRNOTAVAIL;
- if (!otx2_hw_set_mac_addr(pfvf, addr->sa_data))
+ if (!otx2_hw_set_mac_addr(pfvf, addr->sa_data)) {
memcpy(netdev->dev_addr, addr->sa_data, netdev->addr_len);
- else
+ /* update dmac field in vlan offload rule */
+ if (pfvf->flags & OTX2_FLAG_RX_VLAN_SUPPORT)
+ otx2_install_rxvlan_offload_flow(pfvf);
+ } else {
return -EPERM;
+ }
return 0;
}
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index f959688e14a3..7819f278b99c 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -240,10 +240,13 @@ struct otx2_flow_config {
u32 nr_flows;
#define OTX2_MAX_NTUPLE_FLOWS 32
#define OTX2_MAX_UNICAST_FLOWS 8
+#define OTX2_MAX_VLAN_FLOWS 1
#define OTX2_MCAM_COUNT (OTX2_MAX_NTUPLE_FLOWS + \
- OTX2_MAX_UNICAST_FLOWS)
+ OTX2_MAX_UNICAST_FLOWS + \
+ OTX2_MAX_VLAN_FLOWS)
u32 ntuple_offset;
u32 unicast_offset;
+ u32 rx_vlan_offset;
u32 ntuple_max_flows;
struct list_head flow_list;
};
@@ -261,6 +264,7 @@ struct otx2_nic {
#define OTX2_FLAG_MCAM_ENTRIES_ALLOC BIT_ULL(3)
#define OTX2_FLAG_NTUPLE_SUPPORT BIT_ULL(4)
#define OTX2_FLAG_UCAST_FLTR_SUPPORT BIT_ULL(5)
+#define OTX2_FLAG_RX_VLAN_SUPPORT BIT_ULL(6)
#define OTX2_FLAG_RX_PAUSE_ENABLED BIT_ULL(9)
#define OTX2_FLAG_TX_PAUSE_ENABLED BIT_ULL(10)
u64 flags;
@@ -687,5 +691,7 @@ int otx2_prepare_flow_request(struct ethtool_rx_flow_spec *fsp,
struct npc_install_flow_req *req);
int otx2_del_macfilter(struct net_device *netdev, const u8 *mac);
int otx2_add_macfilter(struct net_device *netdev, const u8 *mac);
+int otx2_enable_rxvlan(struct otx2_nic *pf, bool enable);
+int otx2_install_rxvlan_offload_flow(struct otx2_nic *pfvf);
#endif /* OTX2_COMMON_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c
index 7da80aa3f063..7571002316ef 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c
@@ -58,8 +58,11 @@ int otx2_alloc_mcam_entries(struct otx2_nic *pfvf)
flow_cfg->ntuple_offset = 0;
flow_cfg->unicast_offset = flow_cfg->ntuple_offset +
OTX2_MAX_NTUPLE_FLOWS;
+ flow_cfg->rx_vlan_offset = flow_cfg->unicast_offset +
+ OTX2_MAX_UNICAST_FLOWS;
pfvf->flags |= OTX2_FLAG_NTUPLE_SUPPORT;
pfvf->flags |= OTX2_FLAG_UCAST_FLTR_SUPPORT;
+ pfvf->flags |= OTX2_FLAG_RX_VLAN_SUPPORT;
}
for (i = 0; i < rsp->count; i++)
@@ -711,3 +714,102 @@ int otx2_destroy_mcam_flows(struct otx2_nic *pfvf)
return 0;
}
+
+int otx2_install_rxvlan_offload_flow(struct otx2_nic *pfvf)
+{
+ struct otx2_flow_config *flow_cfg = pfvf->flow_cfg;
+ struct npc_install_flow_req *req;
+ int err;
+
+ mutex_lock(&pfvf->mbox.lock);
+ req = otx2_mbox_alloc_msg_npc_install_flow(&pfvf->mbox);
+ if (!req) {
+ mutex_unlock(&pfvf->mbox.lock);
+ return -ENOMEM;
+ }
+
+ req->entry = flow_cfg->entry[flow_cfg->rx_vlan_offset];
+ req->intf = NIX_INTF_RX;
+ ether_addr_copy(req->packet.dmac, pfvf->netdev->dev_addr);
+ eth_broadcast_addr((u8 *)&req->mask.dmac);
+ req->channel = pfvf->hw.rx_chan_base;
+ req->op = NIX_RX_ACTION_DEFAULT;
+ req->features = BIT_ULL(NPC_OUTER_VID) | BIT_ULL(NPC_DMAC);
+ req->vtag0_valid = true;
+ req->vtag0_type = NIX_AF_LFX_RX_VTAG_TYPE0;
+
+ /* Send message to AF */
+ err = otx2_sync_mbox_msg(&pfvf->mbox);
+ mutex_unlock(&pfvf->mbox.lock);
+ return err;
+}
+
+static int otx2_delete_rxvlan_offload_flow(struct otx2_nic *pfvf)
+{
+ struct otx2_flow_config *flow_cfg = pfvf->flow_cfg;
+ struct npc_delete_flow_req *req;
+ int err;
+
+ mutex_lock(&pfvf->mbox.lock);
+ req = otx2_mbox_alloc_msg_npc_delete_flow(&pfvf->mbox);
+ if (!req) {
+ mutex_unlock(&pfvf->mbox.lock);
+ return -ENOMEM;
+ }
+
+ req->entry = flow_cfg->entry[flow_cfg->rx_vlan_offset];
+ /* Send message to AF */
+ err = otx2_sync_mbox_msg(&pfvf->mbox);
+ mutex_unlock(&pfvf->mbox.lock);
+ return err;
+}
+
+int otx2_enable_rxvlan(struct otx2_nic *pf, bool enable)
+{
+ struct nix_vtag_config *req;
+ struct mbox_msghdr *rsp_hdr;
+ int err;
+
+ /* Dont have enough mcam entries */
+ if (!(pf->flags & OTX2_FLAG_RX_VLAN_SUPPORT))
+ return -ENOMEM;
+
+ if (enable) {
+ err = otx2_install_rxvlan_offload_flow(pf);
+ if (err)
+ return err;
+ } else {
+ err = otx2_delete_rxvlan_offload_flow(pf);
+ if (err)
+ return err;
+ }
+
+ mutex_lock(&pf->mbox.lock);
+ req = otx2_mbox_alloc_msg_nix_vtag_cfg(&pf->mbox);
+ if (!req) {
+ mutex_unlock(&pf->mbox.lock);
+ return -ENOMEM;
+ }
+
+ /* config strip, capture and size */
+ req->vtag_size = VTAGSIZE_T4;
+ req->cfg_type = 1; /* rx vlan cfg */
+ req->rx.vtag_type = NIX_AF_LFX_RX_VTAG_TYPE0;
+ req->rx.strip_vtag = enable;
+ req->rx.capture_vtag = enable;
+
+ err = otx2_sync_mbox_msg(&pf->mbox);
+ if (err) {
+ mutex_unlock(&pf->mbox.lock);
+ return err;
+ }
+
+ rsp_hdr = otx2_mbox_get_rsp(&pf->mbox.mbox, 0, &req->hdr);
+ if (IS_ERR(rsp_hdr)) {
+ mutex_unlock(&pf->mbox.lock);
+ return PTR_ERR(rsp_hdr);
+ }
+
+ mutex_unlock(&pf->mbox.lock);
+ return rsp_hdr->rc;
+}
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index ac90302b1d81..22513e58f7bb 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -1566,6 +1566,9 @@ int otx2_open(struct net_device *netdev)
otx2_set_cints_affinity(pf);
+ if (pf->flags & OTX2_FLAG_RX_VLAN_SUPPORT)
+ otx2_enable_rxvlan(pf, true);
+
/* When reinitializing enable time stamping if it is enabled before */
if (pf->flags & OTX2_FLAG_TX_TSTAMP_ENABLED) {
pf->flags &= ~OTX2_FLAG_TX_TSTAMP_ENABLED;
@@ -1763,6 +1766,10 @@ static int otx2_set_features(struct net_device *netdev,
return otx2_cgx_config_loopback(pf,
features & NETIF_F_LOOPBACK);
+ if ((changed & NETIF_F_HW_VLAN_CTAG_RX) && netif_running(netdev))
+ return otx2_enable_rxvlan(pf,
+ features & NETIF_F_HW_VLAN_CTAG_RX);
+
if ((changed & NETIF_F_NTUPLE) && !ntuple)
otx2_destroy_ntuple_flows(pf);
@@ -2138,6 +2145,15 @@ static int otx2_probe(struct pci_dev *pdev, const struct pci_device_id *id)
if (pf->flags & OTX2_FLAG_UCAST_FLTR_SUPPORT)
netdev->priv_flags |= IFF_UNICAST_FLT;
+ /* Support TSO on tag interface */
+ netdev->vlan_features |= netdev->features;
+ netdev->hw_features |= NETIF_F_HW_VLAN_CTAG_TX |
+ NETIF_F_HW_VLAN_STAG_TX;
+ if (pf->flags & OTX2_FLAG_RX_VLAN_SUPPORT)
+ netdev->hw_features |= NETIF_F_HW_VLAN_CTAG_RX |
+ NETIF_F_HW_VLAN_STAG_RX;
+ netdev->features |= netdev->hw_features;
+
netdev->gso_max_segs = OTX2_MAX_GSO_SEGS;
netdev->watchdog_timeo = OTX2_TX_TIMEOUT;
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
index d5d7a2f37493..d0e25414f1a1 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_txrx.c
@@ -556,6 +556,19 @@ static void otx2_sqe_add_ext(struct otx2_nic *pfvf, struct otx2_snd_queue *sq,
ext->tstmp = 1;
}
+#define OTX2_VLAN_PTR_OFFSET (ETH_HLEN - ETH_TLEN)
+ if (skb_vlan_tag_present(skb)) {
+ if (skb->vlan_proto == htons(ETH_P_8021Q)) {
+ ext->vlan1_ins_ena = 1;
+ ext->vlan1_ins_ptr = OTX2_VLAN_PTR_OFFSET;
+ ext->vlan1_ins_tci = skb_vlan_tag_get(skb);
+ } else if (skb->vlan_proto == htons(ETH_P_8021AD)) {
+ ext->vlan0_ins_ena = 1;
+ ext->vlan0_ins_ptr = OTX2_VLAN_PTR_OFFSET;
+ ext->vlan0_ins_tci = skb_vlan_tag_get(skb);
+ }
+ }
+
*offset += sizeof(*ext);
}
@@ -871,6 +884,9 @@ bool otx2_sq_append_skb(struct net_device *netdev, struct otx2_snd_queue *sq,
}
if (skb_shinfo(skb)->gso_size && !is_hw_tso_supported(pfvf, skb)) {
+ /* Insert vlan tag before giving pkt to tso */
+ if (skb_vlan_tag_present(skb))
+ skb = __vlan_hwaccel_push_inside(skb);
otx2_sq_append_tso(pfvf, sq, skb, qidx);
return true;
}
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c
index 67fabf265fe6..d3e4cfd244e2 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_vf.c
@@ -558,6 +558,11 @@ static int otx2vf_probe(struct pci_dev *pdev, const struct pci_device_id *id)
NETIF_F_SG | NETIF_F_TSO | NETIF_F_TSO6 |
NETIF_F_GSO_UDP_L4;
netdev->features = netdev->hw_features;
+ /* Support TSO on tag interface */
+ netdev->vlan_features |= netdev->features;
+ netdev->hw_features |= NETIF_F_HW_VLAN_CTAG_TX |
+ NETIF_F_HW_VLAN_STAG_TX;
+ netdev->features |= netdev->hw_features;
netdev->gso_max_segs = OTX2_MAX_GSO_SEGS;
netdev->watchdog_timeo = OTX2_TX_TIMEOUT;
--
2.16.5
^ permalink raw reply related [flat|nested] 24+ messages in thread* [PATCH v3 net-next 10/13] octeontx2-pf: Add support for SR-IOV management functions
2020-11-11 7:13 [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Naveen Mamindlapalli
` (8 preceding siblings ...)
2020-11-11 7:14 ` [PATCH v3 net-next 09/13] octeontx2-pf: Implement ingress/egress VLAN offload Naveen Mamindlapalli
@ 2020-11-11 7:14 ` Naveen Mamindlapalli
2020-11-11 7:14 ` [PATCH v3 net-next 11/13] octeontx2-af: Handle PF-VF mac address changes Naveen Mamindlapalli
` (3 subsequent siblings)
13 siblings, 0 replies; 24+ messages in thread
From: Naveen Mamindlapalli @ 2020-11-11 7:14 UTC (permalink / raw)
To: netdev, linux-kernel
Cc: kuba, davem, saeed, sgoutham, lcherian, gakula, jerinj, sbhatta,
hkelam
This patch adds support for ndo_set_vf_mac, ndo_set_vf_vlan
and ndo_get_vf_config handlers. The traffic redirection
based on the VF mac address or vlan id is done by installing
MCAM rules. Reserved RX_VTAG_TYPE7 in each NIXLF for VF VLAN
which strips the VLAN tag from ingress VLAN traffic. The NIX PF
allocates two MCAM entries for VF VLAN feature, one used for
ingress VTAG strip and another entry for egress VTAG insertion.
This patch also updates the MAC address in PF installed VF VLAN
rule upon receiving nix_lf_start_rx mbox request for VF since
Administrative Function driver will assign a valid MAC addr
in nix_lf_start_rx function.
Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
Co-developed-by: Tomasz Duszynski <tduszynski@marvell.com>
Signed-off-by: Tomasz Duszynski <tduszynski@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: Hariprasad Kelam <hkelam@marvell.com>
---
drivers/net/ethernet/marvell/octeontx2/af/npc.h | 1 +
drivers/net/ethernet/marvell/octeontx2/af/rvu.h | 4 +
.../net/ethernet/marvell/octeontx2/af/rvu_nix.c | 14 ++
.../net/ethernet/marvell/octeontx2/af/rvu_npc.c | 88 ++++++++
.../net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c | 40 ++++
.../net/ethernet/marvell/octeontx2/af/rvu_struct.h | 11 +
.../ethernet/marvell/octeontx2/nic/otx2_common.h | 9 +
.../ethernet/marvell/octeontx2/nic/otx2_flows.c | 9 +-
.../net/ethernet/marvell/octeontx2/nic/otx2_pf.c | 246 +++++++++++++++++++++
9 files changed, 420 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/npc.h b/drivers/net/ethernet/marvell/octeontx2/af/npc.h
index 860706033aad..35d0348a9d29 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/npc.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/npc.h
@@ -452,6 +452,7 @@ struct rvu_npc_mcam_rule {
bool has_cntr;
u8 default_rule;
bool enable;
+ bool vfvlan_cfg;
};
#endif /* NPC_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
index 6a45573724b4..15e668a87aa1 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
@@ -209,6 +209,7 @@ struct rvu_pfvf {
u16 maxlen;
u16 minlen;
+ u8 pf_set_vf_cfg;
u8 mac_addr[ETH_ALEN]; /* MAC address of this PF/VF */
/* Broadcast pkt replication info */
@@ -596,6 +597,9 @@ void npc_mcam_enable_flows(struct rvu *rvu, u16 target);
void npc_mcam_disable_flows(struct rvu *rvu, u16 target);
void npc_enable_mcam_entry(struct rvu *rvu, struct npc_mcam *mcam,
int blkaddr, int index, bool enable);
+void npc_read_mcam_entry(struct rvu *rvu, struct npc_mcam *mcam,
+ int blkaddr, u16 src, struct mcam_entry *entry,
+ u8 *intf, u8 *ena);
#ifdef CONFIG_DEBUG_FS
void rvu_dbg_init(struct rvu *rvu);
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
index 97a8f932d1e2..53c5556d7061 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
@@ -1196,6 +1196,11 @@ int rvu_mbox_handler_nix_lf_alloc(struct rvu *rvu,
/* Disable NPC entries as NIXLF's contexts are not initialized yet */
rvu_npc_disable_default_entries(rvu, pcifunc, nixlf);
+ /* Configure RX VTAG Type 7 (strip) for vf vlan */
+ rvu_write64(rvu, blkaddr,
+ NIX_AF_LFX_RX_VTAG_TYPEX(nixlf, NIX_AF_LFX_RX_VTAG_TYPE7),
+ VTAGSIZE_T4 | VTAG_STRIP);
+
goto exit;
free_mem:
@@ -1990,6 +1995,10 @@ static int nix_rx_vtag_cfg(struct rvu *rvu, int nixlf, int blkaddr,
req->vtag_size > VTAGSIZE_T8)
return -EINVAL;
+ /* RX VTAG Type 7 reserved for vf vlan */
+ if (req->rx.vtag_type == NIX_AF_LFX_RX_VTAG_TYPE7)
+ return NIX_AF_ERR_RX_VTAG_INUSE;
+
if (req->rx.capture_vtag)
regval |= BIT_ULL(5);
if (req->rx.strip_vtag)
@@ -2933,6 +2942,7 @@ int rvu_mbox_handler_nix_set_mac_addr(struct rvu *rvu,
struct nix_set_mac_addr *req,
struct msg_rsp *rsp)
{
+ bool from_vf = req->hdr.pcifunc & RVU_PFVF_FUNC_MASK;
u16 pcifunc = req->hdr.pcifunc;
int blkaddr, nixlf, err;
struct rvu_pfvf *pfvf;
@@ -2943,6 +2953,10 @@ int rvu_mbox_handler_nix_set_mac_addr(struct rvu *rvu,
pfvf = rvu_get_pfvf(rvu, pcifunc);
+ /* VF can't overwrite admin(PF) changes */
+ if (from_vf && pfvf->pf_set_vf_cfg)
+ return -EPERM;
+
ether_addr_copy(pfvf->mac_addr, req->mac_addr);
rvu_npc_install_ucast_entry(rvu, pcifunc, nixlf,
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
index a7759ecfa586..84e954dafe85 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
@@ -359,6 +359,58 @@ static void npc_get_keyword(struct mcam_entry *entry, int idx,
*cam0 = ~*cam1 & kw_mask;
}
+static void npc_fill_entryword(struct mcam_entry *entry, int idx,
+ u64 cam0, u64 cam1)
+{
+ /* Similar to npc_get_keyword, but fills mcam_entry structure from
+ * CAM registers.
+ */
+ switch (idx) {
+ case 0:
+ entry->kw[0] = cam1;
+ entry->kw_mask[0] = cam1 ^ cam0;
+ break;
+ case 1:
+ entry->kw[1] = cam1;
+ entry->kw_mask[1] = cam1 ^ cam0;
+ break;
+ case 2:
+ entry->kw[1] |= (cam1 & CAM_MASK(16)) << 48;
+ entry->kw[2] = (cam1 >> 16) & CAM_MASK(48);
+ entry->kw_mask[1] |= ((cam1 ^ cam0) & CAM_MASK(16)) << 48;
+ entry->kw_mask[2] = ((cam1 ^ cam0) >> 16) & CAM_MASK(48);
+ break;
+ case 3:
+ entry->kw[2] |= (cam1 & CAM_MASK(16)) << 48;
+ entry->kw[3] = (cam1 >> 16) & CAM_MASK(32);
+ entry->kw_mask[2] |= ((cam1 ^ cam0) & CAM_MASK(16)) << 48;
+ entry->kw_mask[3] = ((cam1 ^ cam0) >> 16) & CAM_MASK(32);
+ break;
+ case 4:
+ entry->kw[3] |= (cam1 & CAM_MASK(32)) << 32;
+ entry->kw[4] = (cam1 >> 32) & CAM_MASK(32);
+ entry->kw_mask[3] |= ((cam1 ^ cam0) & CAM_MASK(32)) << 32;
+ entry->kw_mask[4] = ((cam1 ^ cam0) >> 32) & CAM_MASK(32);
+ break;
+ case 5:
+ entry->kw[4] |= (cam1 & CAM_MASK(32)) << 32;
+ entry->kw[5] = (cam1 >> 32) & CAM_MASK(16);
+ entry->kw_mask[4] |= ((cam1 ^ cam0) & CAM_MASK(32)) << 32;
+ entry->kw_mask[5] = ((cam1 ^ cam0) >> 32) & CAM_MASK(16);
+ break;
+ case 6:
+ entry->kw[5] |= (cam1 & CAM_MASK(48)) << 16;
+ entry->kw[6] = (cam1 >> 48) & CAM_MASK(16);
+ entry->kw_mask[5] |= ((cam1 ^ cam0) & CAM_MASK(48)) << 16;
+ entry->kw_mask[6] = ((cam1 ^ cam0) >> 48) & CAM_MASK(16);
+ break;
+ case 7:
+ entry->kw[6] |= (cam1 & CAM_MASK(48)) << 16;
+ entry->kw_mask[6] |= ((cam1 ^ cam0) & CAM_MASK(48)) << 16;
+ break;
+ }
+}
+
static void npc_get_default_entry_action(struct rvu *rvu, struct npc_mcam *mcam,
int blkaddr, int index,
struct mcam_entry *entry)
@@ -459,6 +511,42 @@ static void npc_config_mcam_entry(struct rvu *rvu, struct npc_mcam *mcam,
npc_enable_mcam_entry(rvu, mcam, blkaddr, actindex, true);
}
+void npc_read_mcam_entry(struct rvu *rvu, struct npc_mcam *mcam,
+ int blkaddr, u16 src,
+ struct mcam_entry *entry, u8 *intf, u8 *ena)
+{
+ int sbank = npc_get_bank(mcam, src);
+ int bank, kw = 0;
+ u64 cam0, cam1;
+
+ src &= (mcam->banksize - 1);
+ bank = sbank;
+
+ for (; bank < (sbank + mcam->banks_per_entry); bank++, kw = kw + 2) {
+ cam1 = rvu_read64(rvu, blkaddr,
+ NPC_AF_MCAMEX_BANKX_CAMX_W0(src, bank, 1));
+ cam0 = rvu_read64(rvu, blkaddr,
+ NPC_AF_MCAMEX_BANKX_CAMX_W0(src, bank, 0));
+ npc_fill_entryword(entry, kw, cam0, cam1);
+
+ cam1 = rvu_read64(rvu, blkaddr,
+ NPC_AF_MCAMEX_BANKX_CAMX_W1(src, bank, 1));
+ cam0 = rvu_read64(rvu, blkaddr,
+ NPC_AF_MCAMEX_BANKX_CAMX_W1(src, bank, 0));
+ npc_fill_entryword(entry, kw + 1, cam0, cam1);
+ }
+
+ entry->action = rvu_read64(rvu, blkaddr,
+ NPC_AF_MCAMEX_BANKX_ACTION(src, sbank));
+ entry->vtag_action =
+ rvu_read64(rvu, blkaddr,
+ NPC_AF_MCAMEX_BANKX_TAG_ACT(src, sbank));
+ *intf = rvu_read64(rvu, blkaddr,
+ NPC_AF_MCAMEX_BANKX_CAMX_INTF(src, sbank, 1)) & 3;
+ *ena = rvu_read64(rvu, blkaddr,
+ NPC_AF_MCAMEX_BANKX_CFG(src, sbank)) & 1;
+}
+
static void npc_copy_mcam_entry(struct rvu *rvu, struct npc_mcam *mcam,
int blkaddr, u16 src, u16 dest)
{
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c
index 94b52dfb6ac6..7bb5e6760509 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c
@@ -1065,6 +1065,9 @@ static int npc_install_flow(struct rvu *rvu, int blkaddr, u16 target,
if (req->default_rule)
pfvf->def_ucast_rule = rule;
+ if (pfvf->pf_set_vf_cfg && req->vtag0_type == NIX_AF_LFX_RX_VTAG_TYPE7)
+ rule->vfvlan_cfg = true;
+
return 0;
}
@@ -1113,6 +1116,10 @@ int rvu_mbox_handler_npc_install_flow(struct rvu *rvu,
pfvf = rvu_get_pfvf(rvu, target);
+ /* PF installing for its VF */
+ if (req->hdr.pcifunc && !from_vf && req->vf)
+ pfvf->pf_set_vf_cfg = 1;
+
/* update req destination mac addr */
if ((req->features & BIT_ULL(NPC_DMAC)) && is_npc_intf_rx(req->intf) &&
is_zero_ether_addr(req->packet.dmac)) {
@@ -1210,6 +1217,36 @@ int rvu_mbox_handler_npc_delete_flow(struct rvu *rvu,
return 0;
}
+static int npc_update_dmac_value(struct rvu *rvu, int npcblkaddr,
+ struct rvu_npc_mcam_rule *rule,
+ struct rvu_pfvf *pfvf)
+{
+ struct npc_mcam_write_entry_req write_req = { 0 };
+ struct mcam_entry *entry = &write_req.entry_data;
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ struct msg_rsp rsp;
+ u8 intf, enable;
+ int err;
+
+ ether_addr_copy(rule->packet.dmac, pfvf->mac_addr);
+
+ npc_read_mcam_entry(rvu, mcam, npcblkaddr, rule->entry,
+ entry, &intf, &enable);
+
+ npc_update_entry(rvu, NPC_DMAC, entry,
+ ether_addr_to_u64(pfvf->mac_addr), 0,
+ 0xffffffffffffull, 0, intf);
+
+ write_req.hdr.pcifunc = rule->owner;
+ write_req.entry = rule->entry;
+
+ mutex_unlock(&mcam->lock);
+ err = rvu_mbox_handler_npc_mcam_write_entry(rvu, &write_req, &rsp);
+ mutex_lock(&mcam->lock);
+
+ return err;
+}
+
void npc_mcam_enable_flows(struct rvu *rvu, u16 target)
{
struct rvu_pfvf *pfvf = rvu_get_pfvf(rvu, target);
@@ -1236,6 +1273,9 @@ void npc_mcam_enable_flows(struct rvu *rvu, u16 target)
continue;
}
+ if (rule->vfvlan_cfg)
+ npc_update_dmac_value(rvu, blkaddr, rule, pfvf);
+
if (rule->rx_action.op == NIX_RX_ACTION_DEFAULT) {
if (!def_ucast_rule)
continue;
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_struct.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu_struct.h
index 9a7eb074cdc2..723643868589 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_struct.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_struct.h
@@ -921,4 +921,15 @@ enum nix_vtag_size {
VTAGSIZE_T4 = 0x0,
VTAGSIZE_T8 = 0x1,
};
+
+enum nix_tx_vtag_op {
+ NOP = 0x0,
+ VTAG_INSERT = 0x1,
+ VTAG_REPLACE = 0x2,
+};
+
+/* NIX RX VTAG actions */
+#define VTAG_STRIP BIT_ULL(4)
+#define VTAG_CAPTURE BIT_ULL(5)
+
#endif /* RVU_STRUCT_H */
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
index 7819f278b99c..b18b45d02165 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.h
@@ -206,6 +206,9 @@ struct otx2_vf_config {
struct otx2_nic *pf;
struct delayed_work link_event_work;
bool intf_down; /* interface was either configured or not */
+ u8 mac[ETH_ALEN];
+ u16 vlan;
+ int tx_vtag_idx;
};
struct flr_work {
@@ -247,6 +250,10 @@ struct otx2_flow_config {
u32 ntuple_offset;
u32 unicast_offset;
u32 rx_vlan_offset;
+ u32 vf_vlan_offset;
+#define OTX2_PER_VF_VLAN_FLOWS 2 /* rx+tx per VF */
+#define OTX2_VF_VLAN_RX_INDEX 0
+#define OTX2_VF_VLAN_TX_INDEX 1
u32 ntuple_max_flows;
struct list_head flow_list;
};
@@ -265,6 +272,8 @@ struct otx2_nic {
#define OTX2_FLAG_NTUPLE_SUPPORT BIT_ULL(4)
#define OTX2_FLAG_UCAST_FLTR_SUPPORT BIT_ULL(5)
#define OTX2_FLAG_RX_VLAN_SUPPORT BIT_ULL(6)
+#define OTX2_FLAG_VF_VLAN_SUPPORT BIT_ULL(7)
+#define OTX2_FLAG_PF_SHUTDOWN BIT_ULL(8)
#define OTX2_FLAG_RX_PAUSE_ENABLED BIT_ULL(9)
#define OTX2_FLAG_TX_PAUSE_ENABLED BIT_ULL(10)
u64 flags;
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c
index 7571002316ef..5b25d2242b8d 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_flows.c
@@ -24,6 +24,7 @@ int otx2_alloc_mcam_entries(struct otx2_nic *pfvf)
struct otx2_flow_config *flow_cfg = pfvf->flow_cfg;
struct npc_mcam_alloc_entry_req *req;
struct npc_mcam_alloc_entry_rsp *rsp;
+ int vf_vlan_max_flows;
int i;
mutex_lock(&pfvf->mbox.lock);
@@ -34,8 +35,9 @@ int otx2_alloc_mcam_entries(struct otx2_nic *pfvf)
return -ENOMEM;
}
+ vf_vlan_max_flows = pfvf->total_vfs * OTX2_PER_VF_VLAN_FLOWS;
req->contig = false;
- req->count = OTX2_MCAM_COUNT;
+ req->count = OTX2_MCAM_COUNT + vf_vlan_max_flows;
/* Send message to AF */
if (otx2_sync_mbox_msg(&pfvf->mbox)) {
@@ -55,7 +57,9 @@ int otx2_alloc_mcam_entries(struct otx2_nic *pfvf)
flow_cfg->ntuple_offset = 0;
pfvf->flags |= OTX2_FLAG_NTUPLE_SUPPORT;
} else {
- flow_cfg->ntuple_offset = 0;
+ flow_cfg->vf_vlan_offset = 0;
+ flow_cfg->ntuple_offset = flow_cfg->vf_vlan_offset +
+ vf_vlan_max_flows;
flow_cfg->unicast_offset = flow_cfg->ntuple_offset +
OTX2_MAX_NTUPLE_FLOWS;
flow_cfg->rx_vlan_offset = flow_cfg->unicast_offset +
@@ -63,6 +67,7 @@ int otx2_alloc_mcam_entries(struct otx2_nic *pfvf)
pfvf->flags |= OTX2_FLAG_NTUPLE_SUPPORT;
pfvf->flags |= OTX2_FLAG_UCAST_FLTR_SUPPORT;
pfvf->flags |= OTX2_FLAG_RX_VLAN_SUPPORT;
+ pfvf->flags |= OTX2_FLAG_VF_VLAN_SUPPORT;
}
for (i = 0; i < rsp->count; i++)
diff --git a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
index 22513e58f7bb..4c82f60f3cf3 100644
--- a/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
+++ b/drivers/net/ethernet/marvell/octeontx2/nic/otx2_pf.c
@@ -1425,6 +1425,8 @@ static void otx2_free_hw_resources(struct otx2_nic *pf)
free_req = otx2_mbox_alloc_msg_nix_lf_free(mbox);
if (free_req) {
free_req->flags = NIX_LF_DISABLE_FLOWS;
+ if (!(pf->flags & OTX2_FLAG_PF_SHUTDOWN))
+ free_req->flags |= NIX_LF_DONT_FREE_TX_VTAG;
if (otx2_sync_mbox_msg(mbox))
dev_err(pf->dev, "%s failed to free nixlf\n", __func__);
}
@@ -1928,6 +1930,245 @@ static int otx2_ioctl(struct net_device *netdev, struct ifreq *req, int cmd)
}
}
+static int otx2_do_set_vf_mac(struct otx2_nic *pf, int vf, const u8 *mac)
+{
+ struct npc_install_flow_req *req;
+ int err;
+
+ mutex_lock(&pf->mbox.lock);
+ req = otx2_mbox_alloc_msg_npc_install_flow(&pf->mbox);
+ if (!req) {
+ err = -ENOMEM;
+ goto out;
+ }
+
+ ether_addr_copy(req->packet.dmac, mac);
+ eth_broadcast_addr((u8 *)&req->mask.dmac);
+ req->features = BIT_ULL(NPC_DMAC);
+ req->channel = pf->hw.rx_chan_base;
+ req->intf = NIX_INTF_RX;
+ req->default_rule = 1;
+ req->append = 1;
+ req->vf = vf + 1;
+ req->op = NIX_RX_ACTION_DEFAULT;
+
+ err = otx2_sync_mbox_msg(&pf->mbox);
+out:
+ mutex_unlock(&pf->mbox.lock);
+ return err;
+}
+
+static int otx2_set_vf_mac(struct net_device *netdev, int vf, u8 *mac)
+{
+ struct otx2_nic *pf = netdev_priv(netdev);
+ struct pci_dev *pdev = pf->pdev;
+ struct otx2_vf_config *config;
+ int ret;
+
+ if (!netif_running(netdev))
+ return -EAGAIN;
+
+ if (vf >= pci_num_vf(pdev))
+ return -EINVAL;
+
+ if (!is_valid_ether_addr(mac))
+ return -EINVAL;
+
+ config = &pf->vf_configs[vf];
+ ether_addr_copy(config->mac, mac);
+
+ ret = otx2_do_set_vf_mac(pf, vf, mac);
+ if (ret == 0)
+ dev_info(&pdev->dev, "Reload VF driver to apply the changes\n");
+
+ return ret;
+}
+
+static int otx2_do_set_vf_vlan(struct otx2_nic *pf, int vf, u16 vlan, u8 qos,
+ __be16 proto)
+{
+ struct otx2_flow_config *flow_cfg = pf->flow_cfg;
+ struct nix_vtag_config_rsp *vtag_rsp;
+ struct npc_delete_flow_req *del_req;
+ struct nix_vtag_config *vtag_req;
+ struct npc_install_flow_req *req;
+ struct otx2_vf_config *config;
+ int err = 0;
+ u32 idx;
+
+ config = &pf->vf_configs[vf];
+
+ if (!vlan && !config->vlan)
+ goto out;
+
+ mutex_lock(&pf->mbox.lock);
+
+ /* free old tx vtag entry */
+ if (config->vlan) {
+ vtag_req = otx2_mbox_alloc_msg_nix_vtag_cfg(&pf->mbox);
+ if (!vtag_req) {
+ err = -ENOMEM;
+ goto out;
+ }
+ vtag_req->cfg_type = 0;
+ vtag_req->tx.free_vtag0 = 1;
+ vtag_req->tx.vtag0_idx = config->tx_vtag_idx;
+
+ err = otx2_sync_mbox_msg(&pf->mbox);
+ if (err)
+ goto out;
+ }
+
+ if (!vlan && config->vlan) {
+ /* rx */
+ del_req = otx2_mbox_alloc_msg_npc_delete_flow(&pf->mbox);
+ if (!del_req) {
+ err = -ENOMEM;
+ goto out;
+ }
+ idx = ((vf * OTX2_PER_VF_VLAN_FLOWS) + OTX2_VF_VLAN_RX_INDEX);
+ del_req->entry =
+ flow_cfg->entry[flow_cfg->vf_vlan_offset + idx];
+ err = otx2_sync_mbox_msg(&pf->mbox);
+ if (err)
+ goto out;
+
+ /* tx */
+ del_req = otx2_mbox_alloc_msg_npc_delete_flow(&pf->mbox);
+ if (!del_req) {
+ err = -ENOMEM;
+ goto out;
+ }
+ idx = ((vf * OTX2_PER_VF_VLAN_FLOWS) + OTX2_VF_VLAN_TX_INDEX);
+ del_req->entry =
+ flow_cfg->entry[flow_cfg->vf_vlan_offset + idx];
+ err = otx2_sync_mbox_msg(&pf->mbox);
+
+ goto out;
+ }
+
+ /* rx */
+ req = otx2_mbox_alloc_msg_npc_install_flow(&pf->mbox);
+ if (!req) {
+ err = -ENOMEM;
+ goto out;
+ }
+
+ idx = ((vf * OTX2_PER_VF_VLAN_FLOWS) + OTX2_VF_VLAN_RX_INDEX);
+ req->entry = flow_cfg->entry[flow_cfg->vf_vlan_offset + idx];
+ req->packet.vlan_tci = htons(vlan);
+ req->mask.vlan_tci = htons(VLAN_VID_MASK);
+ /* af fills the destination mac addr */
+ eth_broadcast_addr((u8 *)&req->mask.dmac);
+ req->features = BIT_ULL(NPC_OUTER_VID) | BIT_ULL(NPC_DMAC);
+ req->channel = pf->hw.rx_chan_base;
+ req->intf = NIX_INTF_RX;
+ req->vf = vf + 1;
+ req->op = NIX_RX_ACTION_DEFAULT;
+ req->vtag0_valid = true;
+ req->vtag0_type = NIX_AF_LFX_RX_VTAG_TYPE7;
+ req->set_cntr = 1;
+
+ err = otx2_sync_mbox_msg(&pf->mbox);
+ if (err)
+ goto out;
+
+ /* tx */
+ vtag_req = otx2_mbox_alloc_msg_nix_vtag_cfg(&pf->mbox);
+ if (!vtag_req) {
+ err = -ENOMEM;
+ goto out;
+ }
+
+ /* configure tx vtag params */
+ vtag_req->vtag_size = VTAGSIZE_T4;
+ vtag_req->cfg_type = 0; /* tx vlan cfg */
+ vtag_req->tx.cfg_vtag0 = 1;
+ vtag_req->tx.vtag0 = (ntohs(proto) << 16) | vlan;
+
+ err = otx2_sync_mbox_msg(&pf->mbox);
+ if (err)
+ goto out;
+
+ vtag_rsp = (struct nix_vtag_config_rsp *)otx2_mbox_get_rsp
+ (&pf->mbox.mbox, 0, &vtag_req->hdr);
+ if (IS_ERR(vtag_rsp)) {
+ err = PTR_ERR(vtag_rsp);
+ goto out;
+ }
+ config->tx_vtag_idx = vtag_rsp->vtag0_idx;
+
+ req = otx2_mbox_alloc_msg_npc_install_flow(&pf->mbox);
+ if (!req) {
+ err = -ENOMEM;
+ goto out;
+ }
+
+ eth_zero_addr((u8 *)&req->mask.dmac);
+ idx = ((vf * OTX2_PER_VF_VLAN_FLOWS) + OTX2_VF_VLAN_TX_INDEX);
+ req->entry = flow_cfg->entry[flow_cfg->vf_vlan_offset + idx];
+ req->features = BIT_ULL(NPC_DMAC);
+ req->channel = pf->hw.tx_chan_base;
+ req->intf = NIX_INTF_TX;
+ req->vf = vf + 1;
+ req->op = NIX_TX_ACTIONOP_UCAST_DEFAULT;
+ req->vtag0_def = vtag_rsp->vtag0_idx;
+ req->vtag0_op = VTAG_INSERT;
+ req->set_cntr = 1;
+
+ err = otx2_sync_mbox_msg(&pf->mbox);
+out:
+ config->vlan = vlan;
+ mutex_unlock(&pf->mbox.lock);
+ return err;
+}
+
+static int otx2_set_vf_vlan(struct net_device *netdev, int vf, u16 vlan, u8 qos,
+ __be16 proto)
+{
+ struct otx2_nic *pf = netdev_priv(netdev);
+ struct pci_dev *pdev = pf->pdev;
+
+ if (!netif_running(netdev))
+ return -EAGAIN;
+
+ if (vf >= pci_num_vf(pdev))
+ return -EINVAL;
+
+ /* qos is currently unsupported */
+ if (vlan >= VLAN_N_VID || qos)
+ return -EINVAL;
+
+ if (proto != htons(ETH_P_8021Q))
+ return -EPROTONOSUPPORT;
+
+ if (!(pf->flags & OTX2_FLAG_VF_VLAN_SUPPORT))
+ return -EOPNOTSUPP;
+
+ return otx2_do_set_vf_vlan(pf, vf, vlan, qos, proto);
+}
+
+static int otx2_get_vf_config(struct net_device *netdev, int vf,
+ struct ifla_vf_info *ivi)
+{
+ struct otx2_nic *pf = netdev_priv(netdev);
+ struct pci_dev *pdev = pf->pdev;
+ struct otx2_vf_config *config;
+
+ if (!netif_running(netdev))
+ return -EAGAIN;
+
+ if (vf >= pci_num_vf(pdev))
+ return -EINVAL;
+
+ config = &pf->vf_configs[vf];
+ ivi->vf = vf;
+ ether_addr_copy(ivi->mac, config->mac);
+ ivi->vlan = config->vlan;
+
+ return 0;
+}
+
static const struct net_device_ops otx2_netdev_ops = {
.ndo_open = otx2_open,
.ndo_stop = otx2_stop,
@@ -1939,6 +2180,9 @@ static const struct net_device_ops otx2_netdev_ops = {
.ndo_tx_timeout = otx2_tx_timeout,
.ndo_get_stats64 = otx2_get_stats64,
.ndo_do_ioctl = otx2_ioctl,
+ .ndo_set_vf_mac = otx2_set_vf_mac,
+ .ndo_set_vf_vlan = otx2_set_vf_vlan,
+ .ndo_get_vf_config = otx2_get_vf_config,
};
static int otx2_wq_init(struct otx2_nic *pf)
@@ -2331,6 +2575,8 @@ static void otx2_remove(struct pci_dev *pdev)
pf = netdev_priv(netdev);
+ pf->flags |= OTX2_FLAG_PF_SHUTDOWN;
+
if (pf->flags & OTX2_FLAG_TX_TSTAMP_ENABLED)
otx2_config_hw_tx_tstamp(pf, false);
if (pf->flags & OTX2_FLAG_RX_TSTAMP_ENABLED)
--
2.16.5
^ permalink raw reply related [flat|nested] 24+ messages in thread* [PATCH v3 net-next 11/13] octeontx2-af: Handle PF-VF mac address changes
2020-11-11 7:13 [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Naveen Mamindlapalli
` (9 preceding siblings ...)
2020-11-11 7:14 ` [PATCH v3 net-next 10/13] octeontx2-pf: Add support for SR-IOV management functions Naveen Mamindlapalli
@ 2020-11-11 7:14 ` Naveen Mamindlapalli
2020-11-11 7:14 ` [PATCH v3 net-next 12/13] octeontx2-af: Add new mbox messages to retrieve MCAM entries Naveen Mamindlapalli
` (2 subsequent siblings)
13 siblings, 0 replies; 24+ messages in thread
From: Naveen Mamindlapalli @ 2020-11-11 7:14 UTC (permalink / raw)
To: netdev, linux-kernel
Cc: kuba, davem, saeed, sgoutham, lcherian, gakula, jerinj, sbhatta,
hkelam
From: Hariprasad Kelam <hkelam@marvell.com>
This patch handles the VF mac address changes as given below.
1. mac addr configrued by VF will be retained until VF module unload.
2. mac addr configred by PF for VF will be retained until power cycle.
3. mac addr confgired by PF for its VF can't be overwritten by VF.
Signed-off-by: Hariprasad Kelam <hkelam@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
---
drivers/net/ethernet/marvell/octeontx2/af/rvu.c | 12 +++++++++++-
drivers/net/ethernet/marvell/octeontx2/af/rvu.h | 2 ++
drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c | 9 +++++++++
drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c | 17 ++++++++++++++---
4 files changed, 36 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu.c
index e8b5aaf73201..9f901c0edcbb 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.c
@@ -727,6 +727,10 @@ static void rvu_setup_pfvf_macaddress(struct rvu *rvu)
u64 *mac;
for (pf = 0; pf < hw->total_pfs; pf++) {
+ /* For PF0(AF), Assign MAC address to only VFs (LBKVFs) */
+ if (!pf)
+ goto lbkvf;
+
if (!is_pf_cgxmapped(rvu, pf))
continue;
/* Assign MAC address to PF */
@@ -740,8 +744,10 @@ static void rvu_setup_pfvf_macaddress(struct rvu *rvu)
} else {
eth_random_addr(pfvf->mac_addr);
}
+ ether_addr_copy(pfvf->default_mac, pfvf->mac_addr);
- /* Assign MAC address to VFs */
+lbkvf:
+ /* Assign MAC address to VFs*/
rvu_get_pf_numvfs(rvu, pf, &numvfs, &hwvf);
for (vf = 0; vf < numvfs; vf++, hwvf++) {
pfvf = &rvu->hwvf[hwvf];
@@ -754,6 +760,7 @@ static void rvu_setup_pfvf_macaddress(struct rvu *rvu)
} else {
eth_random_addr(pfvf->mac_addr);
}
+ ether_addr_copy(pfvf->default_mac, pfvf->mac_addr);
}
}
}
@@ -1176,6 +1183,9 @@ static void rvu_detach_block(struct rvu *rvu, int pcifunc, int blktype)
if (blkaddr < 0)
return;
+ if (blktype == BLKTYPE_NIX)
+ rvu_nix_reset_mac(pfvf, pcifunc);
+
block = &hw->block[blkaddr];
num_lfs = rvu_get_rsrc_mapcount(pfvf, block->addr);
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
index 15e668a87aa1..4bc75be6d45b 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
@@ -211,6 +211,7 @@ struct rvu_pfvf {
u8 pf_set_vf_cfg;
u8 mac_addr[ETH_ALEN]; /* MAC address of this PF/VF */
+ u8 default_mac[ETH_ALEN]; /* MAC address from FWdata */
/* Broadcast pkt replication info */
u16 bcast_mce_idx;
@@ -553,6 +554,7 @@ int nix_get_nixlf(struct rvu *rvu, u16 pcifunc, int *nixlf, int *nix_blkaddr);
int nix_update_bcast_mce_list(struct rvu *rvu, u16 pcifunc, bool add);
struct nix_hw *get_nix_hw(struct rvu_hwinfo *hw, int blkaddr);
int rvu_get_next_nix_blkaddr(struct rvu *rvu, int blkaddr);
+void rvu_nix_reset_mac(struct rvu_pfvf *pfvf, int pcifunc);
/* NPC APIs */
int rvu_npc_init(struct rvu *rvu);
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
index 53c5556d7061..f5affb809c77 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
@@ -3737,3 +3737,12 @@ int rvu_mbox_handler_nix_lso_format_cfg(struct rvu *rvu,
return 0;
}
+
+void rvu_nix_reset_mac(struct rvu_pfvf *pfvf, int pcifunc)
+{
+ bool from_vf = !!(pcifunc & RVU_PFVF_FUNC_MASK);
+
+ /* overwrite vf mac address with default_mac */
+ if (from_vf)
+ ether_addr_copy(pfvf->mac_addr, pfvf->default_mac);
+}
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c
index 7bb5e6760509..2f4030821d8d 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc_fs.c
@@ -938,7 +938,8 @@ static void npc_update_tx_entry(struct rvu *rvu, struct rvu_pfvf *pfvf,
static int npc_install_flow(struct rvu *rvu, int blkaddr, u16 target,
int nixlf, struct rvu_pfvf *pfvf,
struct npc_install_flow_req *req,
- struct npc_install_flow_rsp *rsp, bool enable)
+ struct npc_install_flow_rsp *rsp, bool enable,
+ bool pf_set_vfs_mac)
{
struct rvu_npc_mcam_rule *def_ucast_rule = pfvf->def_ucast_rule;
u64 features, installed_features, missing_features = 0;
@@ -1065,6 +1066,12 @@ static int npc_install_flow(struct rvu *rvu, int blkaddr, u16 target,
if (req->default_rule)
pfvf->def_ucast_rule = rule;
+ /* VF's MAC address is being changed via PF */
+ if (pf_set_vfs_mac) {
+ ether_addr_copy(pfvf->default_mac, req->packet.dmac);
+ ether_addr_copy(pfvf->mac_addr, req->packet.dmac);
+ }
+
if (pfvf->pf_set_vf_cfg && req->vtag0_type == NIX_AF_LFX_RX_VTAG_TYPE7)
rule->vfvlan_cfg = true;
@@ -1078,6 +1085,7 @@ int rvu_mbox_handler_npc_install_flow(struct rvu *rvu,
bool from_vf = !!(req->hdr.pcifunc & RVU_PFVF_FUNC_MASK);
int blkaddr, nixlf, err;
struct rvu_pfvf *pfvf;
+ bool pf_set_vfs_mac = false;
bool enable = true;
u16 target;
@@ -1102,8 +1110,11 @@ int rvu_mbox_handler_npc_install_flow(struct rvu *rvu,
if (!req->hdr.pcifunc)
target = req->vf;
/* PF installing for its VF */
- else if (!from_vf && req->vf)
+ else if (!from_vf && req->vf) {
target = (req->hdr.pcifunc & ~RVU_PFVF_FUNC_MASK) | req->vf;
+ pf_set_vfs_mac = req->default_rule &&
+ (req->features & BIT_ULL(NPC_DMAC));
+ }
/* msg received from PF/VF */
else
target = req->hdr.pcifunc;
@@ -1152,7 +1163,7 @@ int rvu_mbox_handler_npc_install_flow(struct rvu *rvu,
return -EINVAL;
return npc_install_flow(rvu, blkaddr, target, nixlf, pfvf, req, rsp,
- enable);
+ enable, pf_set_vfs_mac);
}
static int npc_delete_flow(struct rvu *rvu, struct rvu_npc_mcam_rule *rule,
--
2.16.5
^ permalink raw reply related [flat|nested] 24+ messages in thread* [PATCH v3 net-next 12/13] octeontx2-af: Add new mbox messages to retrieve MCAM entries
2020-11-11 7:13 [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Naveen Mamindlapalli
` (10 preceding siblings ...)
2020-11-11 7:14 ` [PATCH v3 net-next 11/13] octeontx2-af: Handle PF-VF mac address changes Naveen Mamindlapalli
@ 2020-11-11 7:14 ` Naveen Mamindlapalli
2020-11-11 7:14 ` [PATCH v3 net-next 13/13] octeontx2-af: Delete NIX_RXVLAN_ALLOC mailbox message Naveen Mamindlapalli
2020-11-12 20:16 ` [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Saeed Mahameed
13 siblings, 0 replies; 24+ messages in thread
From: Naveen Mamindlapalli @ 2020-11-11 7:14 UTC (permalink / raw)
To: netdev, linux-kernel
Cc: kuba, davem, saeed, sgoutham, lcherian, gakula, jerinj, sbhatta,
hkelam
This patch introduces new mailbox mesages to retrieve a given
MCAM entry or base flow steering rule of a VF installed by its
parent PF. This helps while updating the existing MCAM rules
with out re-framing the whole mailbox request again. The INSTALL
FLOW mailbox consumer can read-modify-write the existing entry.
Similarly while installing new flow rules for a VF, the base
flow steering rule match creteria is copied to the new flow rule
and the deltas are appended to the new rule.
Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
Co-developed-by: Vamsi Attunuru <vattunuru@marvell.com>
Signed-off-by: Vamsi Attunuru <vattunuru@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
---
drivers/net/ethernet/marvell/octeontx2/af/mbox.h | 22 +++++++
.../net/ethernet/marvell/octeontx2/af/rvu_npc.c | 70 ++++++++++++++++++++++
2 files changed, 92 insertions(+)
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
index ef20078508a5..8ea132ec1784 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
@@ -192,6 +192,11 @@ M(NPC_INSTALL_FLOW, 0x600d, npc_install_flow, \
npc_install_flow_req, npc_install_flow_rsp) \
M(NPC_DELETE_FLOW, 0x600e, npc_delete_flow, \
npc_delete_flow_req, msg_rsp) \
+M(NPC_MCAM_READ_ENTRY, 0x600f, npc_mcam_read_entry, \
+ npc_mcam_read_entry_req, \
+ npc_mcam_read_entry_rsp) \
+M(NPC_MCAM_READ_BASE_RULE, 0x6011, npc_read_base_steer_rule, \
+ msg_req, npc_mcam_read_base_rule_rsp) \
/* NIX mbox IDs (range 0x8000 - 0xFFFF) */ \
M(NIX_LF_ALLOC, 0x8000, nix_lf_alloc, \
nix_lf_alloc_req, nix_lf_alloc_rsp) \
@@ -1009,6 +1014,23 @@ struct npc_delete_flow_req {
u8 all; /* PF + VFs */
};
+struct npc_mcam_read_entry_req {
+ struct mbox_msghdr hdr;
+ u16 entry; /* MCAM entry to read */
+};
+
+struct npc_mcam_read_entry_rsp {
+ struct mbox_msghdr hdr;
+ struct mcam_entry entry_data;
+ u8 intf;
+ u8 enable;
+};
+
+struct npc_mcam_read_base_rule_rsp {
+ struct mbox_msghdr hdr;
+ struct mcam_entry entry;
+};
+
enum ptp_op {
PTP_OP_ADJFINE = 0,
PTP_OP_GET_CLOCK = 1,
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
index 84e954dafe85..9b5c5b376cc6 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
@@ -2194,6 +2194,30 @@ int rvu_mbox_handler_npc_mcam_free_entry(struct rvu *rvu,
return rc;
}
+int rvu_mbox_handler_npc_mcam_read_entry(struct rvu *rvu,
+ struct npc_mcam_read_entry_req *req,
+ struct npc_mcam_read_entry_rsp *rsp)
+{
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ u16 pcifunc = req->hdr.pcifunc;
+ int blkaddr, rc;
+
+ blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NPC, 0);
+ if (blkaddr < 0)
+ return NPC_MCAM_INVALID_REQ;
+
+ mutex_lock(&mcam->lock);
+ rc = npc_mcam_verify_entry(mcam, pcifunc, req->entry);
+ if (!rc) {
+ npc_read_mcam_entry(rvu, mcam, blkaddr, req->entry,
+ &rsp->entry_data,
+ &rsp->intf, &rsp->enable);
+ }
+
+ mutex_unlock(&mcam->lock);
+ return rc;
+}
+
int rvu_mbox_handler_npc_mcam_write_entry(struct rvu *rvu,
struct npc_mcam_write_entry_req *req,
struct msg_rsp *rsp)
@@ -2754,3 +2778,49 @@ bool rvu_npc_write_default_rule(struct rvu *rvu, int blkaddr, int nixlf,
return enable;
}
+
+int rvu_mbox_handler_npc_read_base_steer_rule(struct rvu *rvu,
+ struct msg_req *req,
+ struct npc_mcam_read_base_rule_rsp *rsp)
+{
+ struct npc_mcam *mcam = &rvu->hw->mcam;
+ int index, blkaddr, nixlf, rc = 0;
+ u16 pcifunc = req->hdr.pcifunc;
+ struct rvu_pfvf *pfvf;
+ u8 intf, enable;
+
+ blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NPC, 0);
+ if (blkaddr < 0)
+ return NPC_MCAM_INVALID_REQ;
+
+ /* Return the channel number in case of PF */
+ if (!(pcifunc & RVU_PFVF_FUNC_MASK)) {
+ pfvf = rvu_get_pfvf(rvu, pcifunc);
+ rsp->entry.kw[0] = pfvf->rx_chan_base;
+ rsp->entry.kw_mask[0] = 0xFFFULL;
+ goto out;
+ }
+
+ /* Find the pkt steering rule installed by PF to this VF */
+ mutex_lock(&mcam->lock);
+ for (index = 0; index < mcam->bmap_entries; index++) {
+ if (mcam->entry2target_pffunc[index] == pcifunc)
+ goto read_entry;
+ }
+
+ rc = nix_get_nixlf(rvu, pcifunc, &nixlf, NULL);
+ if (rc < 0) {
+ mutex_unlock(&mcam->lock);
+ goto out;
+ }
+ /* Read the default ucast entry if there is no pkt steering rule */
+ index = npc_get_nixlf_mcam_index(mcam, pcifunc, nixlf,
+ NIXLF_UCAST_ENTRY);
+read_entry:
+ /* Read the mcam entry */
+ npc_read_mcam_entry(rvu, mcam, blkaddr, index, &rsp->entry, &intf,
+ &enable);
+ mutex_unlock(&mcam->lock);
+out:
+ return rc;
+}
--
2.16.5
^ permalink raw reply related [flat|nested] 24+ messages in thread* [PATCH v3 net-next 13/13] octeontx2-af: Delete NIX_RXVLAN_ALLOC mailbox message
2020-11-11 7:13 [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Naveen Mamindlapalli
` (11 preceding siblings ...)
2020-11-11 7:14 ` [PATCH v3 net-next 12/13] octeontx2-af: Add new mbox messages to retrieve MCAM entries Naveen Mamindlapalli
@ 2020-11-11 7:14 ` Naveen Mamindlapalli
2020-11-12 20:16 ` [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Saeed Mahameed
13 siblings, 0 replies; 24+ messages in thread
From: Naveen Mamindlapalli @ 2020-11-11 7:14 UTC (permalink / raw)
To: netdev, linux-kernel
Cc: kuba, davem, saeed, sgoutham, lcherian, gakula, jerinj, sbhatta,
hkelam
From: Subbaraya Sundeep <sbhatta@marvell.com>
Since mailbox message for installing flows is in place,
remove the RXVLAN_ALLOC mbox message which is redundant.
Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: Naveen Mamindlapalli <naveenm@marvell.com>
---
drivers/net/ethernet/marvell/octeontx2/af/mbox.h | 1 -
drivers/net/ethernet/marvell/octeontx2/af/rvu.h | 6 ---
.../net/ethernet/marvell/octeontx2/af/rvu_nix.c | 59 ----------------------
.../net/ethernet/marvell/octeontx2/af/rvu_npc.c | 24 ---------
4 files changed, 90 deletions(-)
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
index 8ea132ec1784..cb4e3d86b58b 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/mbox.h
@@ -226,7 +226,6 @@ M(NIX_SET_RX_CFG, 0x8010, nix_set_rx_cfg, nix_rx_cfg, msg_rsp) \
M(NIX_LSO_FORMAT_CFG, 0x8011, nix_lso_format_cfg, \
nix_lso_format_cfg, \
nix_lso_format_cfg_rsp) \
-M(NIX_RXVLAN_ALLOC, 0x8012, nix_rxvlan_alloc, msg_req, msg_rsp) \
M(NIX_LF_PTP_TX_ENABLE, 0x8013, nix_lf_ptp_tx_enable, msg_req, msg_rsp) \
M(NIX_LF_PTP_TX_DISABLE, 0x8014, nix_lf_ptp_tx_disable, msg_req, msg_rsp) \
M(NIX_BP_ENABLE, 0x8016, nix_bp_enable, nix_bp_cfg_req, \
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
index 4bc75be6d45b..044dde24865c 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu.h
@@ -217,11 +217,6 @@ struct rvu_pfvf {
u16 bcast_mce_idx;
struct nix_mce_list bcast_mce_list;
- /* VLAN offload */
- struct mcam_entry entry;
- int rxvlan_index;
- bool rxvlan;
-
struct rvu_npc_mcam_rule *def_ucast_rule;
bool cgx_in_use; /* this PF/VF using CGX? */
@@ -571,7 +566,6 @@ void rvu_npc_enable_promisc_entry(struct rvu *rvu, u16 pcifunc, int nixlf);
void rvu_npc_install_bcast_match_entry(struct rvu *rvu, u16 pcifunc,
int nixlf, u64 chan);
void rvu_npc_enable_bcast_entry(struct rvu *rvu, u16 pcifunc, bool enable);
-int rvu_npc_update_rxvlan(struct rvu *rvu, u16 pcifunc, int nixlf);
void rvu_npc_disable_mcam_entries(struct rvu *rvu, u16 pcifunc, int nixlf);
void rvu_npc_free_mcam_entries(struct rvu *rvu, u16 pcifunc, int nixlf);
void rvu_npc_disable_default_entries(struct rvu *rvu, u16 pcifunc, int nixlf);
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
index f5affb809c77..89601e796c85 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_nix.c
@@ -3144,65 +3144,6 @@ int rvu_mbox_handler_nix_set_hw_frs(struct rvu *rvu, struct nix_frs_cfg *req,
return 0;
}
-int rvu_mbox_handler_nix_rxvlan_alloc(struct rvu *rvu, struct msg_req *req,
- struct msg_rsp *rsp)
-{
- struct npc_mcam_alloc_entry_req alloc_req = { };
- struct npc_mcam_alloc_entry_rsp alloc_rsp = { };
- struct npc_mcam_free_entry_req free_req = { };
- u16 pcifunc = req->hdr.pcifunc;
- int blkaddr, nixlf, err;
- struct rvu_pfvf *pfvf;
-
- /* LBK VFs do not have separate MCAM UCAST entry hence
- * skip allocating rxvlan for them
- */
- if (is_afvf(pcifunc))
- return 0;
-
- pfvf = rvu_get_pfvf(rvu, pcifunc);
- if (pfvf->rxvlan)
- return 0;
-
- /* alloc new mcam entry */
- alloc_req.hdr.pcifunc = pcifunc;
- alloc_req.count = 1;
-
- err = rvu_mbox_handler_npc_mcam_alloc_entry(rvu, &alloc_req,
- &alloc_rsp);
- if (err)
- return err;
-
- /* update entry to enable rxvlan offload */
- blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NIX, pcifunc);
- if (blkaddr < 0) {
- err = NIX_AF_ERR_AF_LF_INVALID;
- goto free_entry;
- }
-
- nixlf = rvu_get_lf(rvu, &rvu->hw->block[blkaddr], pcifunc, 0);
- if (nixlf < 0) {
- err = NIX_AF_ERR_AF_LF_INVALID;
- goto free_entry;
- }
-
- pfvf->rxvlan_index = alloc_rsp.entry_list[0];
- /* all it means is that rxvlan_index is valid */
- pfvf->rxvlan = true;
-
- err = rvu_npc_update_rxvlan(rvu, pcifunc, nixlf);
- if (err)
- goto free_entry;
-
- return 0;
-free_entry:
- free_req.hdr.pcifunc = pcifunc;
- free_req.entry = alloc_rsp.entry_list[0];
- rvu_mbox_handler_npc_mcam_free_entry(rvu, &free_req, rsp);
- pfvf->rxvlan = false;
- return err;
-}
-
int rvu_mbox_handler_nix_set_rx_cfg(struct rvu *rvu, struct nix_rx_cfg *req,
struct msg_rsp *rsp)
{
diff --git a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
index 9b5c5b376cc6..a4764f8c1c39 100644
--- a/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
+++ b/drivers/net/ethernet/marvell/octeontx2/af/rvu_npc.c
@@ -2731,30 +2731,6 @@ int rvu_mbox_handler_npc_get_kex_cfg(struct rvu *rvu, struct msg_req *req,
return 0;
}
-int rvu_npc_update_rxvlan(struct rvu *rvu, u16 pcifunc, int nixlf)
-{
- struct rvu_pfvf *pfvf = rvu_get_pfvf(rvu, pcifunc);
- struct npc_mcam *mcam = &rvu->hw->mcam;
- int blkaddr, index;
- bool enable;
-
- blkaddr = rvu_get_blkaddr(rvu, BLKTYPE_NPC, 0);
- if (blkaddr < 0)
- return NIX_AF_ERR_AF_LF_INVALID;
-
- if (!pfvf->rxvlan)
- return 0;
-
- index = npc_get_nixlf_mcam_index(mcam, pcifunc, nixlf,
- NIXLF_UCAST_ENTRY);
- pfvf->entry.action = npc_get_mcam_action(rvu, mcam, blkaddr, index);
- enable = is_mcam_entry_enabled(rvu, mcam, blkaddr, index);
- npc_config_mcam_entry(rvu, mcam, blkaddr, pfvf->rxvlan_index,
- pfvf->nix_rx_intf, &pfvf->entry, enable);
-
- return 0;
-}
-
bool rvu_npc_write_default_rule(struct rvu *rvu, int blkaddr, int nixlf,
u16 pcifunc, u8 intf, struct mcam_entry *entry,
int *index)
--
2.16.5
^ permalink raw reply related [flat|nested] 24+ messages in thread* Re: [PATCH v3 net-next 00/13] Add ethtool ntuple filters support
2020-11-11 7:13 [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Naveen Mamindlapalli
` (12 preceding siblings ...)
2020-11-11 7:14 ` [PATCH v3 net-next 13/13] octeontx2-af: Delete NIX_RXVLAN_ALLOC mailbox message Naveen Mamindlapalli
@ 2020-11-12 20:16 ` Saeed Mahameed
2020-11-14 19:00 ` [EXT] " Naveen Mamindlapalli
13 siblings, 1 reply; 24+ messages in thread
From: Saeed Mahameed @ 2020-11-12 20:16 UTC (permalink / raw)
To: Naveen Mamindlapalli, netdev, linux-kernel
Cc: kuba, davem, sgoutham, lcherian, gakula, jerinj, sbhatta, hkelam
On Wed, 2020-11-11 at 12:43 +0530, Naveen Mamindlapalli wrote:
> This patch series adds support for ethtool ntuple filters, unicast
> address filtering, VLAN offload and SR-IOV ndo handlers. All of the
> above features are based on the Admin Function(AF) driver support to
> install and delete the low level MCAM entries. Each MCAM entry is
> programmed with the packet fields to match and what actions to take
> if the match succeeds. The PF driver requests AF driver to allocate
> set of MCAM entries to be used to install the flows by that PF. The
> entries will be freed when the PF driver is unloaded.
>
> * The patches 1 to 4 adds AF driver infrastructure to install and
> delete the low level MCAM flow entries.
> * Patch 5 adds ethtool ntuple filter support.
> * Patch 6 adds unicast MAC address filtering.
> * Patch 7 adds support for dumping the MCAM entries via debugfs.
> * Patches 8 to 10 adds support for VLAN offload.
> * Patch 10 to 11 adds support for SR-IOV ndo handlers.
> * Patch 12 adds support to read the MCAM entries.
>
> Misc:
> * Removed redundant mailbox NIX_RXVLAN_ALLOC.
>
> Change-log:
> v3:
> - Fixed Saeed's review comments on v2.
> - Fixed modifying the netdev->flags from driver.
> - Fixed modifying the netdev features and hw_features after
> register_netdev.
> - Removed unwanted ndo_features_check callback.
> v2:
> - Fixed the sparse issues reported by Jakub.
>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
^ permalink raw reply [flat|nested] 24+ messages in thread* RE: [EXT] Re: [PATCH v3 net-next 00/13] Add ethtool ntuple filters support
2020-11-12 20:16 ` [PATCH v3 net-next 00/13] Add ethtool ntuple filters support Saeed Mahameed
@ 2020-11-14 19:00 ` Naveen Mamindlapalli
0 siblings, 0 replies; 24+ messages in thread
From: Naveen Mamindlapalli @ 2020-11-14 19:00 UTC (permalink / raw)
To: Saeed Mahameed, netdev@vger.kernel.org,
linux-kernel@vger.kernel.org
Cc: kuba@kernel.org, davem@davemloft.net, Sunil Kovvuri Goutham,
Linu Cherian, Geethasowjanya Akula, Jerin Jacob Kollanukkaran,
Subbaraya Sundeep Bhatta, Hariprasad Kelam
Hi Saeed,
Thanks for the review.
> -----Original Message-----
> From: Saeed Mahameed <saeed@kernel.org>
> Sent: Friday, November 13, 2020 1:47 AM
> To: Naveen Mamindlapalli <naveenm@marvell.com>; netdev@vger.kernel.org;
> linux-kernel@vger.kernel.org
> Cc: kuba@kernel.org; davem@davemloft.net; Sunil Kovvuri Goutham
> <sgoutham@marvell.com>; Linu Cherian <lcherian@marvell.com>;
> Geethasowjanya Akula <gakula@marvell.com>; Jerin Jacob Kollanukkaran
> <jerinj@marvell.com>; Subbaraya Sundeep Bhatta <sbhatta@marvell.com>;
> Hariprasad Kelam <hkelam@marvell.com>
> Subject: [EXT] Re: [PATCH v3 net-next 00/13] Add ethtool ntuple filters support
>
> External Email
>
> ----------------------------------------------------------------------
> On Wed, 2020-11-11 at 12:43 +0530, Naveen Mamindlapalli wrote:
> > This patch series adds support for ethtool ntuple filters, unicast
> > address filtering, VLAN offload and SR-IOV ndo handlers. All of the
> > above features are based on the Admin Function(AF) driver support to
> > install and delete the low level MCAM entries. Each MCAM entry is
> > programmed with the packet fields to match and what actions to take if
> > the match succeeds. The PF driver requests AF driver to allocate set
> > of MCAM entries to be used to install the flows by that PF. The
> > entries will be freed when the PF driver is unloaded.
> >
> > * The patches 1 to 4 adds AF driver infrastructure to install and
> > delete the low level MCAM flow entries.
> > * Patch 5 adds ethtool ntuple filter support.
> > * Patch 6 adds unicast MAC address filtering.
> > * Patch 7 adds support for dumping the MCAM entries via debugfs.
> > * Patches 8 to 10 adds support for VLAN offload.
> > * Patch 10 to 11 adds support for SR-IOV ndo handlers.
> > * Patch 12 adds support to read the MCAM entries.
> >
> > Misc:
> > * Removed redundant mailbox NIX_RXVLAN_ALLOC.
> >
> > Change-log:
> > v3:
> > - Fixed Saeed's review comments on v2.
> > - Fixed modifying the netdev->flags from driver.
> > - Fixed modifying the netdev features and hw_features after
> > register_netdev.
> > - Removed unwanted ndo_features_check callback.
> > v2:
> > - Fixed the sparse issues reported by Jakub.
> >
>
> Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
^ permalink raw reply [flat|nested] 24+ messages in thread