octeontx2-af: cn10k: Add support for programmable channels

NIX uses unique channel numbers to identify the packet sources/sinks
like CGX,LBK and SDP. The channel numbers assigned to each block are
hardwired in CN9xxx silicon.
The fixed channel numbers in CN9xxx are:

0x0 | a << 8 | b            - LBK(0..3)_CH(0..63)
0x0 | a << 8                - Reserved
0x700 | a                   - SDP_CH(0..255)
0x800 | a << 8 | b << 4 | c - CGX(0..7)_LMAC(0..3)_CH(0..15)

All the channels in the above fixed enumerator(with maximum
number of blocks) are not required since some chips
have less number of blocks.
For CN10K silicon the channel numbers need to be programmed by
software in each block with the base channel number and range of
channels. This patch calculates and assigns the channel numbers
to efficiently distribute the channel number range(0-4095) among
all the blocks. The assignment is made based on the actual number of
blocks present and also contiguously leaving no holes.
The channel numbers remaining after the math are used as new CPT
replay channels present in CN10K. Also since channel numbers are
not fixed the transmit channel link number needed by AF consumers
is calculated by AF and sent along with nix_lf_alloc mailbox response.

Signed-off-by: Subbaraya Sundeep <sbhatta@marvell.com>
Signed-off-by: Geetha sowjanya <gakula@marvell.com>
Signed-off-by: Sunil Goutham <sgoutham@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

MAINTAINERS

@@ -10720,6 +10720,7 @@ M:	Linu Cherian <lcherian@marvell.com>
 M:	Geetha sowjanya <gakula@marvell.com>
 M:	Jerin Jacob <jerinj@marvell.com>
 M:	hariprasad <hkelam@marvell.com>
+M:	Subbaraya Sundeep <sbhatta@marvell.com>
 L:	netdev@vger.kernel.org
 S:	Supported
 F:	Documentation/networking/device_drivers/ethernet/marvell/octeontx2.rst

drivers/net/ethernet/marvell/octeontx2/af/Makefile

@@ -10,4 +10,4 @@ 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_npc_fs.o \
-		  rvu_cpt.o rvu_devlink.o rpm.o
+		  rvu_cpt.o rvu_devlink.o rpm.o rvu_cn10k.o

drivers/net/ethernet/marvell/octeontx2/af/cgx.c

@@ -123,6 +123,20 @@ void *cgx_get_pdata(int cgx_id)
 	return NULL;
 }
 
+void cgx_lmac_write(int cgx_id, int lmac_id, u64 offset, u64 val)
+{
+	struct cgx *cgx_dev = cgx_get_pdata(cgx_id);
+
+	cgx_write(cgx_dev, lmac_id, offset, val);
+}
+
+u64 cgx_lmac_read(int cgx_id, int lmac_id, u64 offset)
+{
+	struct cgx *cgx_dev = cgx_get_pdata(cgx_id);
+
+	return cgx_read(cgx_dev, lmac_id, offset);
+}
+
 int cgx_get_cgxid(void *cgxd)
 {
 	struct cgx *cgx = cgxd;

drivers/net/ethernet/marvell/octeontx2/af/cgx.h

@@ -162,4 +162,6 @@ struct mac_ops *get_mac_ops(void *cgxd);
 int cgx_get_nr_lmacs(void *cgxd);
 u8 cgx_get_lmacid(void *cgxd, u8 lmac_index);
 unsigned long cgx_get_lmac_bmap(void *cgxd);
+void cgx_lmac_write(int cgx_id, int lmac_id, u64 offset, u64 val);
+u64 cgx_lmac_read(int cgx_id, int lmac_id, u64 offset);
 #endif /* CGX_H */

drivers/net/ethernet/marvell/octeontx2/af/common.h

@@ -191,6 +191,9 @@ enum nix_scheduler {
 #define NIX_LINK_LBK(a)			(12 + (a))
 #define NIX_CHAN_CGX_LMAC_CHX(a, b, c)	(0x800 + 0x100 * (a) + 0x10 * (b) + (c))
 #define NIX_CHAN_LBK_CHX(a, b)		(0 + 0x100 * (a) + (b))
+#define NIX_CHAN_SDP_CH_START		(0x700ull)
+
+#define SDP_CHANNELS			256
 
 /* NIX LSO format indices.
  * As of now TSO is the only one using, so statically assigning indices.

drivers/net/ethernet/marvell/octeontx2/af/rpm.h

@@ -8,13 +8,19 @@
 #ifndef RPM_H
 #define RPM_H
 
+#include <linux/bits.h>
+
 /* PCI device IDs */
 #define PCI_DEVID_CN10K_RPM		0xA060
 
 /* Registers */
 #define RPMX_CMRX_SW_INT                0x180
 #define RPMX_CMRX_SW_INT_W1S            0x188
 #define RPMX_CMRX_SW_INT_ENA_W1S        0x198
+#define RPMX_CMRX_LINK_CFG		0x1070
+
+#define RPMX_CMRX_LINK_RANGE_MASK	GENMASK_ULL(19, 16)
+#define RPMX_CMRX_LINK_BASE_MASK	GENMASK_ULL(11, 0)
 
 #define RPM_LMAC_FWI			0xa
 

drivers/net/ethernet/marvell/octeontx2/af/rvu.c

@@ -1006,6 +1006,10 @@ static int rvu_setup_hw_resources(struct rvu *rvu)
 		rvu_scan_block(rvu, block);
 	}
 
+	err = rvu_set_channels_base(rvu);
+	if (err)
+		goto msix_err;
+
 	err = rvu_npc_init(rvu);
 	if (err)
 		goto npc_err;
@@ -1025,6 +1029,8 @@ static int rvu_setup_hw_resources(struct rvu *rvu)
 	if (err)
 		goto nix_err;
 
+	rvu_program_channels(rvu);
+
 	return 0;
 
 nix_err:
@@ -2721,8 +2727,6 @@ static void rvu_enable_afvf_intr(struct rvu *rvu)
 	rvupf_write64(rvu, RVU_PF_VFME_INT_ENA_W1SX(1), INTR_MASK(vfs - 64));
 }
 
-#define PCI_DEVID_OCTEONTX2_LBK 0xA061
-
 int rvu_get_num_lbk_chans(void)
 {
 	struct pci_dev *pdev;

drivers/net/ethernet/marvell/octeontx2/af/rvu.h

@@ -19,9 +19,11 @@
 #include "common.h"
 #include "mbox.h"
 #include "npc.h"
+#include "rvu_reg.h"
 
 /* PCI device IDs */
 #define	PCI_DEVID_OCTEONTX2_RVU_AF		0xA065
+#define	PCI_DEVID_OCTEONTX2_LBK			0xA061
 
 /* Subsystem Device ID */
 #define PCI_SUBSYS_DEVID_96XX                  0xB200
@@ -305,6 +307,7 @@ struct hw_cap {
 	bool	nix_tx_link_bp;		 /* Can link backpressure TL queues ? */
 	bool	nix_rx_multicast;	 /* Rx packet replication support */
 	bool	per_pf_mbox_regs; /* PF mbox specified in per PF registers ? */
+	bool	programmable_chans; /* Channels programmable ? */
 };
 
 struct rvu_hwinfo {
@@ -313,9 +316,14 @@ struct rvu_hwinfo {
 	u16	max_vfs_per_pf; /* Max VFs that can be attached to a PF */
 	u8	cgx;
 	u8	lmac_per_cgx;
+	u16	cgx_chan_base;	/* CGX base channel number */
+	u16	lbk_chan_base;	/* LBK base channel number */
+	u16	sdp_chan_base;	/* SDP base channel number */
+	u16	cpt_chan_base;	/* CPT base channel number */
 	u8	cgx_links;
 	u8	lbk_links;
 	u8	sdp_links;
+	u8	cpt_links;	/* Number of CPT links */
 	u8	npc_kpus;          /* No of parser units */
 	u8	npc_pkinds;        /* No of port kinds */
 	u8	npc_intfs;         /* No of interfaces */
@@ -499,6 +507,38 @@ static inline bool is_rvu_otx2(struct rvu *rvu)
 		midr == PCI_REVISION_ID_95XXMM);
 }
 
+static inline u16 rvu_nix_chan_cgx(struct rvu *rvu, u8 cgxid,
+				   u8 lmacid, u8 chan)
+{
+	u64 nix_const = rvu_read64(rvu, BLKADDR_NIX0, NIX_AF_CONST);
+	u16 cgx_chans = nix_const & 0xFFULL;
+	struct rvu_hwinfo *hw = rvu->hw;
+
+	if (!hw->cap.programmable_chans)
+		return NIX_CHAN_CGX_LMAC_CHX(cgxid, lmacid, chan);
+
+	return rvu->hw->cgx_chan_base +
+		(cgxid * hw->lmac_per_cgx + lmacid) * cgx_chans + chan;
+}
+
+static inline u16 rvu_nix_chan_lbk(struct rvu *rvu, u8 lbkid,
+				   u8 chan)
+{
+	u64 nix_const = rvu_read64(rvu, BLKADDR_NIX0, NIX_AF_CONST);
+	u16 lbk_chans = (nix_const >> 16) & 0xFFULL;
+	struct rvu_hwinfo *hw = rvu->hw;
+
+	if (!hw->cap.programmable_chans)
+		return NIX_CHAN_LBK_CHX(lbkid, chan);
+
+	return rvu->hw->lbk_chan_base + lbkid * lbk_chans + chan;
+}
+
+static inline u16 rvu_nix_chan_cpt(struct rvu *rvu, u8 chan)
+{
+	return rvu->hw->cpt_chan_base + chan;
+}
+
 /* Function Prototypes
  * RVU
  */
@@ -640,6 +680,10 @@ bool is_mac_feature_supported(struct rvu *rvu, int pf, int feature);
 /* CPT APIs */
 int rvu_cpt_lf_teardown(struct rvu *rvu, u16 pcifunc, int lf, int slot);
 
+/* CN10K RVU */
+int rvu_set_channels_base(struct rvu *rvu);
+void rvu_program_channels(struct rvu *rvu);
+
 #ifdef CONFIG_DEBUG_FS
 void rvu_dbg_init(struct rvu *rvu);
 void rvu_dbg_exit(struct rvu *rvu);