diff --git a/Documentation/networking/ti-cpsw.txt b/Documentation/networking/ti-cpsw.txt
new file mode 100644
index 0000000000000000000000000000000000000000..67039205bd6961c6c3afbf64f8d29450bee737f7
--- /dev/null
+++ b/Documentation/networking/ti-cpsw.txt
@@ -0,0 +1,540 @@
+* Texas Instruments CPSW ethernet driver
+
+Multiqueue & CBS & MQPRIO
+=====================================================================
+=====================================================================
+
+The cpsw has 3 CBS shapers for each external ports. This document
+describes MQPRIO and CBS Qdisc offload configuration for cpsw driver
+based on examples. It potentially can be used in audio video bridging
+(AVB) and time sensitive networking (TSN).
+
+The following examples were tested on AM572x EVM and BBB boards.
+
+Test setup
+==========
+
+Under consideration two examples with AM572x EVM running cpsw driver
+in dual_emac mode.
+
+Several prerequisites:
+- TX queues must be rated starting from txq0 that has highest priority
+- Traffic classes are used starting from 0, that has highest priority
+- CBS shapers should be used with rated queues
+- The bandwidth for CBS shapers has to be set a little bit more then
+ potential incoming rate, thus, rate of all incoming tx queues has
+ to be a little less
+- Real rates can differ, due to discreetness
+- Map skb-priority to txq is not enough, also skb-priority to l2 prio
+ map has to be created with ip or vconfig tool
+- Any l2/socket prio (0 - 7) for classes can be used, but for
+ simplicity default values are used: 3 and 2
+- only 2 classes tested: A and B, but checked and can work with more,
+ maximum allowed 4, but only for 3 rate can be set.
+
+Test setup for examples
+=======================
+ +-------------------------------+
+ |--+ |
+ | | Workstation0 |
+ |E | MAC 18:03:73:66:87:42 |
++-----------------------------+ +--|t | |
+| | 1 | E | | |h |./tsn_listener -d \ |
+| Target board: | 0 | t |--+ |0 | 18:03:73:66:87:42 -i eth0 \|
+| AM572x EVM | 0 | h | | | -s 1500 |
+| | 0 | 0 | |--+ |
+| Only 2 classes: |Mb +---| +-------------------------------+
+| class A, class B | |
+| | +---| +-------------------------------+
+| | 1 | E | |--+ |
+| | 0 | t | | | Workstation1 |
+| | 0 | h |--+ |E | MAC 20:cf:30:85:7d:fd |
+| |Mb | 1 | +--|t | |
++-----------------------------+ |h |./tsn_listener -d \ |
+ |0 | 20:cf:30:85:7d:fd -i eth0 \|
+ | | -s 1500 |
+ |--+ |
+ +-------------------------------+
+
+*********************************************************************
+*********************************************************************
+*********************************************************************
+Example 1: One port tx AVB configuration scheme for target board
+----------------------------------------------------------------------
+(prints and scheme for AM572x evm, applicable for single port boards)
+
+tc - traffic class
+txq - transmit queue
+p - priority
+f - fifo (cpsw fifo)
+S - shaper configured
+
++------------------------------------------------------------------+ u
+| +---------------+ +---------------+ +------+ +------+ | s
+| | | | | | | | | | e
+| | App 1 | | App 2 | | Apps | | Apps | | r
+| | Class A | | Class B | | Rest | | Rest | |
+| | Eth0 | | Eth0 | | Eth0 | | Eth1 | | s
+| | VLAN100 | | VLAN100 | | | | | | | | p
+| | 40 Mb/s | | 20 Mb/s | | | | | | | | a
+| | SO_PRIORITY=3 | | SO_PRIORITY=2 | | | | | | | | c
+| | | | | | | | | | | | | | e
+| +---|-----------+ +---|-----------+ +---|--+ +---|--+ |
++-----|------------------|------------------|--------|-------------+
+ +-+ +------------+ | |
+ | | +-----------------+ +--+
+ | | | |
++---|-------|-------------|-----------------------|----------------+
+| +----+ +----+ +----+ +----+ +----+ |
+| | p3 | | p2 | | p1 | | p0 | | p0 | | k
+| \ / \ / \ / \ / \ / | e
+| \ / \ / \ / \ / \ / | r
+| \/ \/ \/ \/ \/ | n
+| | | | | | e
+| | | +-----+ | | l
+| | | | | |
+| +----+ +----+ +----+ +----+ | s
+| |tc0 | |tc1 | |tc2 | |tc0 | | p
+| \ / \ / \ / \ / | a
+| \ / \ / \ / \ / | c
+| \/ \/ \/ \/ | e
+| | | +-----+ | |
+| | | | | | |
+| | | | | | |
+| | | | | | |
+| +----+ +----+ +----+ +----+ +----+ |
+| |txq0| |txq1| |txq2| |txq3| |txq4| |
+| \ / \ / \ / \ / \ / |
+| \ / \ / \ / \ / \ / |
+| \/ \/ \/ \/ \/ |
+| +-|------|------|------|--+ +--|--------------+ |
+| | | | | | | Eth0.100 | | Eth1 | |
++---|------|------|------|------------------------|----------------+
+ | | | | |
+ p p p p |
+ 3 2 0-1, 4-7 <- L2 priority |
+ | | | | |
+ | | | | |
++---|------|------|------|------------------------|----------------+
+| | | | | |----------+ |
+| +----+ +----+ +----+ +----+ +----+ |
+| |dma7| |dma6| |dma5| |dma4| |dma3| |
+| \ / \ / \ / \ / \ / | c
+| \S / \S / \ / \ / \ / | p
+| \/ \/ \/ \/ \/ | s
+| | | | +----- | | w
+| | | | | | |
+| | | | | | | d
+| +----+ +----+ +----+p p+----+ | r
+| | | | | | |o o| | | i
+| | f3 | | f2 | | f0 |r r| f0 | | v
+| |tc0 | |tc1 | |tc2 |t t|tc0 | | e
+| \CBS / \CBS / \CBS /1 2\CBS / | r
+| \S / \S / \ / \ / |
+| \/ \/ \/ \/ |
++------------------------------------------------------------------+
+========================================Eth==========================>
+
+1)
+// Add 4 tx queues, for interface Eth0, and 1 tx queue for Eth1
+$ ethtool -L eth0 rx 1 tx 5
+rx unmodified, ignoring
+
+2)
+// Check if num of queues is set correctly:
+$ ethtool -l eth0
+Channel parameters for eth0:
+Pre-set maximums:
+RX: 8
+TX: 8
+Other: 0
+Combined: 0
+Current hardware settings:
+RX: 1
+TX: 5
+Other: 0
+Combined: 0
+
+3)
+// TX queues must be rated starting from 0, so set bws for tx0 and tx1
+// Set rates 40 and 20 Mb/s appropriately.
+// Pay attention, real speed can differ a bit due to discreetness.
+// Leave last 2 tx queues not rated.
+$ echo 40 > /sys/class/net/eth0/queues/tx-0/tx_maxrate
+$ echo 20 > /sys/class/net/eth0/queues/tx-1/tx_maxrate
+
+4)
+// Check maximum rate of tx (cpdma) queues:
+$ cat /sys/class/net/eth0/queues/tx-*/tx_maxrate
+40
+20
+0
+0
+0
+
+5)
+// Map skb->priority to traffic class:
+// 3pri -> tc0, 2pri -> tc1, (0,1,4-7)pri -> tc2
+// Map traffic class to transmit queue:
+// tc0 -> txq0, tc1 -> txq1, tc2 -> (txq2, txq3)
+$ tc qdisc replace dev eth0 handle 100: parent root mqprio num_tc 3 \
+map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 queues 1@0 1@1 2@2 hw 1
+
+5a)
+// As two interface sharing same set of tx queues, assign all traffic
+// coming to interface Eth1 to separate queue in order to not mix it
+// with traffic from interface Eth0, so use separate txq to send
+// packets to Eth1, so all prio -> tc0 and tc0 -> txq4
+// Here hw 0, so here still default configuration for eth1 in hw
+$ tc qdisc replace dev eth1 handle 100: parent root mqprio num_tc 1 \
+map 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 queues 1@4 hw 0
+
+6)
+// Check classes settings
+$ tc -g class show dev eth0
++---(100:ffe2) mqprio
+| +---(100:3) mqprio
+| +---(100:4) mqprio
+|
++---(100:ffe1) mqprio
+| +---(100:2) mqprio
+|
++---(100:ffe0) mqprio
+ +---(100:1) mqprio
+
+$ tc -g class show dev eth1
++---(100:ffe0) mqprio
+ +---(100:5) mqprio
+
+7)
+// Set rate for class A - 41 Mbit (tc0, txq0) using CBS Qdisc
+// Set it +1 Mb for reserve (important!)
+// here only idle slope is important, others arg are ignored
+// Pay attention, real speed can differ a bit due to discreetness
+$ tc qdisc add dev eth0 parent 100:1 cbs locredit -1438 \
+hicredit 62 sendslope -959000 idleslope 41000 offload 1
+net eth0: set FIFO3 bw = 50
+
+8)
+// Set rate for class B - 21 Mbit (tc1, txq1) using CBS Qdisc:
+// Set it +1 Mb for reserve (important!)
+$ tc qdisc add dev eth0 parent 100:2 cbs locredit -1468 \
+hicredit 65 sendslope -979000 idleslope 21000 offload 1
+net eth0: set FIFO2 bw = 30
+
+9)
+// Create vlan 100 to map sk->priority to vlan qos
+$ ip link add link eth0 name eth0.100 type vlan id 100
+8021q: 802.1Q VLAN Support v1.8
+8021q: adding VLAN 0 to HW filter on device eth0
+8021q: adding VLAN 0 to HW filter on device eth1
+net eth0: Adding vlanid 100 to vlan filter
+
+10)
+// Map skb->priority to L2 prio, 1 to 1
+$ ip link set eth0.100 type vlan \
+egress 0:0 1:1 2:2 3:3 4:4 5:5 6:6 7:7
+
+11)
+// Check egress map for vlan 100
+$ cat /proc/net/vlan/eth0.100
+[...]
+INGRESS priority mappings: 0:0 1:0 2:0 3:0 4:0 5:0 6:0 7:0
+EGRESS priority mappings: 0:0 1:1 2:2 3:3 4:4 5:5 6:6 7:7
+
+12)
+// Run your appropriate tools with socket option "SO_PRIORITY"
+// to 3 for class A and/or to 2 for class B
+// (I took at https://www.spinics.net/lists/netdev/msg460869.html)
+./tsn_talker -d 18:03:73:66:87:42 -i eth0.100 -p3 -s 1500&
+./tsn_talker -d 18:03:73:66:87:42 -i eth0.100 -p2 -s 1500&
+
+13)
+// run your listener on workstation (should be in same vlan)
+// (I took at https://www.spinics.net/lists/netdev/msg460869.html)
+./tsn_listener -d 18:03:73:66:87:42 -i enp5s0 -s 1500
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39000 kbps
+
+14)
+// Restore default configuration if needed
+$ ip link del eth0.100
+$ tc qdisc del dev eth1 root
+$ tc qdisc del dev eth0 root
+net eth0: Prev FIFO2 is shaped
+net eth0: set FIFO3 bw = 0
+net eth0: set FIFO2 bw = 0
+$ ethtool -L eth0 rx 1 tx 1
+
+*********************************************************************
+*********************************************************************
+*********************************************************************
+Example 2: Two port tx AVB configuration scheme for target board
+----------------------------------------------------------------------
+(prints and scheme for AM572x evm, for dual emac boards only)
+
++------------------------------------------------------------------+ u
+| +----------+ +----------+ +------+ +----------+ +----------+ | s
+| | | | | | | | | | | | e
+| | App 1 | | App 2 | | Apps | | App 3 | | App 4 | | r
+| | Class A | | Class B | | Rest | | Class B | | Class A | |
+| | Eth0 | | Eth0 | | | | | Eth1 | | Eth1 | | s
+| | VLAN100 | | VLAN100 | | | | | VLAN100 | | VLAN100 | | p
+| | 40 Mb/s | | 20 Mb/s | | | | | 10 Mb/s | | 30 Mb/s | | a
+| | SO_PRI=3 | | SO_PRI=2 | | | | | SO_PRI=3 | | SO_PRI=2 | | c
+| | | | | | | | | | | | | | | | | e
+| +---|------+ +---|------+ +---|--+ +---|------+ +---|------+ |
++-----|-------------|-------------|---------|-------------|--------+
+ +-+ +-------+ | +----------+ +----+
+ | | +-------+------+ | |
+ | | | | | |
++---|-------|-------------|--------------|-------------|-------|---+
+| +----+ +----+ +----+ +----+ +----+ +----+ +----+ +----+ |
+| | p3 | | p2 | | p1 | | p0 | | p0 | | p1 | | p2 | | p3 | | k
+| \ / \ / \ / \ / \ / \ / \ / \ / | e
+| \ / \ / \ / \ / \ / \ / \ / \ / | r
+| \/ \/ \/ \/ \/ \/ \/ \/ | n
+| | | | | | | | e
+| | | +----+ +----+ | | | l
+| | | | | | | |
+| +----+ +----+ +----+ +----+ +----+ +----+ | s
+| |tc0 | |tc1 | |tc2 | |tc2 | |tc1 | |tc0 | | p
+| \ / \ / \ / \ / \ / \ / | a
+| \ / \ / \ / \ / \ / \ / | c
+| \/ \/ \/ \/ \/ \/ | e
+| | | +-----+ +-----+ | | |
+| | | | | | | | | |
+| | | | | | | | | |
+| | | | | E E | | | | |
+| +----+ +----+ +----+ +----+ t t +----+ +----+ +----+ +----+ |
+| |txq0| |txq1| |txq4| |txq5| h h |txq6| |txq7| |txq3| |txq2| |
+| \ / \ / \ / \ / 0 1 \ / \ / \ / \ / |
+| \ / \ / \ / \ / . . \ / \ / \ / \ / |
+| \/ \/ \/ \/ 1 1 \/ \/ \/ \/ |
+| +-|------|------|------|--+ 0 0 +-|------|------|------|--+ |
+| | | | | | | 0 0 | | | | | | |
++---|------|------|------|---------------|------|------|------|----+
+ | | | | | | | |
+ p p p p p p p p
+ 3 2 0-1, 4-7 <-L2 pri-> 0-1, 4-7 2 3
+ | | | | | | | |
+ | | | | | | | |
++---|------|------|------|---------------|------|------|------|----+
+| | | | | | | | | |
+| +----+ +----+ +----+ +----+ +----+ +----+ +----+ +----+ |
+| |dma7| |dma6| |dma3| |dma2| |dma1| |dma0| |dma4| |dma5| |
+| \ / \ / \ / \ / \ / \ / \ / \ / | c
+| \S / \S / \ / \ / \ / \ / \S / \S / | p
+| \/ \/ \/ \/ \/ \/ \/ \/ | s
+| | | | +----- | | | | | w
+| | | | | +----+ | | | |
+| | | | | | | | | | d
+| +----+ +----+ +----+p p+----+ +----+ +----+ | r
+| | | | | | |o o| | | | | | | i
+| | f3 | | f2 | | f0 |r CPSW r| f3 | | f2 | | f0 | | v
+| |tc0 | |tc1 | |tc2 |t t|tc0 | |tc1 | |tc2 | | e
+| \CBS / \CBS / \CBS /1 2\CBS / \CBS / \CBS / | r
+| \S / \S / \ / \S / \S / \ / |
+| \/ \/ \/ \/ \/ \/ |
++------------------------------------------------------------------+
+========================================Eth==========================>
+
+1)
+// Add 8 tx queues, for interface Eth0, but they are common, so are accessed
+// by two interfaces Eth0 and Eth1.
+$ ethtool -L eth1 rx 1 tx 8
+rx unmodified, ignoring
+
+2)
+// Check if num of queues is set correctly:
+$ ethtool -l eth0
+Channel parameters for eth0:
+Pre-set maximums:
+RX: 8
+TX: 8
+Other: 0
+Combined: 0
+Current hardware settings:
+RX: 1
+TX: 8
+Other: 0
+Combined: 0
+
+3)
+// TX queues must be rated starting from 0, so set bws for tx0 and tx1 for Eth0
+// and for tx2 and tx3 for Eth1. That is, rates 40 and 20 Mb/s appropriately
+// for Eth0 and 30 and 10 Mb/s for Eth1.
+// Real speed can differ a bit due to discreetness
+// Leave last 4 tx queues as not rated
+$ echo 40 > /sys/class/net/eth0/queues/tx-0/tx_maxrate
+$ echo 20 > /sys/class/net/eth0/queues/tx-1/tx_maxrate
+$ echo 30 > /sys/class/net/eth1/queues/tx-2/tx_maxrate
+$ echo 10 > /sys/class/net/eth1/queues/tx-3/tx_maxrate
+
+4)
+// Check maximum rate of tx (cpdma) queues:
+$ cat /sys/class/net/eth0/queues/tx-*/tx_maxrate
+40
+20
+30
+10
+0
+0
+0
+0
+
+5)
+// Map skb->priority to traffic class for Eth0:
+// 3pri -> tc0, 2pri -> tc1, (0,1,4-7)pri -> tc2
+// Map traffic class to transmit queue:
+// tc0 -> txq0, tc1 -> txq1, tc2 -> (txq4, txq5)
+$ tc qdisc replace dev eth0 handle 100: parent root mqprio num_tc 3 \
+map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 queues 1@0 1@1 2@4 hw 1
+
+6)
+// Check classes settings
+$ tc -g class show dev eth0
++---(100:ffe2) mqprio
+| +---(100:5) mqprio
+| +---(100:6) mqprio
+|
++---(100:ffe1) mqprio
+| +---(100:2) mqprio
+|
++---(100:ffe0) mqprio
+ +---(100:1) mqprio
+
+7)
+// Set rate for class A - 41 Mbit (tc0, txq0) using CBS Qdisc for Eth0
+// here only idle slope is important, others ignored
+// Real speed can differ a bit due to discreetness
+$ tc qdisc add dev eth0 parent 100:1 cbs locredit -1470 \
+hicredit 62 sendslope -959000 idleslope 41000 offload 1
+net eth0: set FIFO3 bw = 50
+
+8)
+// Set rate for class B - 21 Mbit (tc1, txq1) using CBS Qdisc for Eth0
+$ tc qdisc add dev eth0 parent 100:2 cbs locredit -1470 \
+hicredit 65 sendslope -979000 idleslope 21000 offload 1
+net eth0: set FIFO2 bw = 30
+
+9)
+// Create vlan 100 to map sk->priority to vlan qos for Eth0
+$ ip link add link eth0 name eth0.100 type vlan id 100
+net eth0: Adding vlanid 100 to vlan filter
+
+10)
+// Map skb->priority to L2 prio for Eth0.100, one to one
+$ ip link set eth0.100 type vlan \
+egress 0:0 1:1 2:2 3:3 4:4 5:5 6:6 7:7
+
+11)
+// Check egress map for vlan 100
+$ cat /proc/net/vlan/eth0.100
+[...]
+INGRESS priority mappings: 0:0 1:0 2:0 3:0 4:0 5:0 6:0 7:0
+EGRESS priority mappings: 0:0 1:1 2:2 3:3 4:4 5:5 6:6 7:7
+
+12)
+// Map skb->priority to traffic class for Eth1:
+// 3pri -> tc0, 2pri -> tc1, (0,1,4-7)pri -> tc2
+// Map traffic class to transmit queue:
+// tc0 -> txq2, tc1 -> txq3, tc2 -> (txq6, txq7)
+$ tc qdisc replace dev eth1 handle 100: parent root mqprio num_tc 3 \
+map 2 2 1 0 2 2 2 2 2 2 2 2 2 2 2 2 queues 1@2 1@3 2@6 hw 1
+
+13)
+// Check classes settings
+$ tc -g class show dev eth1
++---(100:ffe2) mqprio
+| +---(100:7) mqprio
+| +---(100:8) mqprio
+|
++---(100:ffe1) mqprio
+| +---(100:4) mqprio
+|
++---(100:ffe0) mqprio
+ +---(100:3) mqprio
+
+14)
+// Set rate for class A - 31 Mbit (tc0, txq2) using CBS Qdisc for Eth1
+// here only idle slope is important, others ignored
+// Set it +1 Mb for reserve (important!)
+$ tc qdisc add dev eth1 parent 100:3 cbs locredit -1453 \
+hicredit 47 sendslope -969000 idleslope 31000 offload 1
+net eth1: set FIFO3 bw = 31
+
+15)
+// Set rate for class B - 11 Mbit (tc1, txq3) using CBS Qdisc for Eth1
+// Set it +1 Mb for reserve (important!)
+$ tc qdisc add dev eth1 parent 100:4 cbs locredit -1483 \
+hicredit 34 sendslope -989000 idleslope 11000 offload 1
+net eth1: set FIFO2 bw = 11
+
+16)
+// Create vlan 100 to map sk->priority to vlan qos for Eth1
+$ ip link add link eth1 name eth1.100 type vlan id 100
+net eth1: Adding vlanid 100 to vlan filter
+
+17)
+// Map skb->priority to L2 prio for Eth1.100, one to one
+$ ip link set eth1.100 type vlan \
+egress 0:0 1:1 2:2 3:3 4:4 5:5 6:6 7:7
+
+18)
+// Check egress map for vlan 100
+$ cat /proc/net/vlan/eth1.100
+[...]
+INGRESS priority mappings: 0:0 1:0 2:0 3:0 4:0 5:0 6:0 7:0
+EGRESS priority mappings: 0:0 1:1 2:2 3:3 4:4 5:5 6:6 7:7
+
+19)
+// Run appropriate tools with socket option "SO_PRIORITY" to 3
+// for class A and to 2 for class B. For both interfaces
+./tsn_talker -d 18:03:73:66:87:42 -i eth0.100 -p2 -s 1500&
+./tsn_talker -d 18:03:73:66:87:42 -i eth0.100 -p3 -s 1500&
+./tsn_talker -d 20:cf:30:85:7d:fd -i eth1.100 -p2 -s 1500&
+./tsn_talker -d 20:cf:30:85:7d:fd -i eth1.100 -p3 -s 1500&
+
+20)
+// run your listener on workstation (should be in same vlan)
+// (I took at https://www.spinics.net/lists/netdev/msg460869.html)
+./tsn_listener -d 18:03:73:66:87:42 -i enp5s0 -s 1500
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39012 kbps
+Receiving data rate: 39000 kbps
+
+21)
+// Restore default configuration if needed
+$ ip link del eth1.100
+$ ip link del eth0.100
+$ tc qdisc del dev eth1 root
+net eth1: Prev FIFO2 is shaped
+net eth1: set FIFO3 bw = 0
+net eth1: set FIFO2 bw = 0
+$ tc qdisc del dev eth0 root
+net eth0: Prev FIFO2 is shaped
+net eth0: set FIFO3 bw = 0
+net eth0: set FIFO2 bw = 0
+$ ethtool -L eth0 rx 1 tx 1
diff --git a/drivers/net/ethernet/ti/cpsw.c b/drivers/net/ethernet/ti/cpsw.c
index 00761fe59848de7ccc90f91f45128ceb61927905..171abcfb618408a9a19ddf809056813605542045 100644
--- a/drivers/net/ethernet/ti/cpsw.c
+++ b/drivers/net/ethernet/ti/cpsw.c
@@ -39,12 +39,15 @@
#include <linux/sys_soc.h>
#include <linux/pinctrl/consumer.h>
+#include <net/pkt_cls.h>
#include "cpsw.h"
#include "cpsw_ale.h"
#include "cpts.h"
#include "davinci_cpdma.h"
+#include <net/pkt_sched.h>
+
#define CPSW_DEBUG (NETIF_MSG_HW | NETIF_MSG_WOL | \
NETIF_MSG_DRV | NETIF_MSG_LINK | \
NETIF_MSG_IFUP | NETIF_MSG_INTR | \
@@ -153,6 +156,12 @@ do { \
#define IRQ_NUM 2
#define CPSW_MAX_QUEUES 8
#define CPSW_CPDMA_DESCS_POOL_SIZE_DEFAULT 256
+#define CPSW_FIFO_QUEUE_TYPE_SHIFT 16
+#define CPSW_FIFO_SHAPE_EN_SHIFT 16
+#define CPSW_FIFO_RATE_EN_SHIFT 20
+#define CPSW_TC_NUM 4
+#define CPSW_FIFO_SHAPERS_NUM (CPSW_TC_NUM - 1)
+#define CPSW_PCT_MASK 0x7f
#define CPSW_RX_VLAN_ENCAP_HDR_PRIO_SHIFT 29
#define CPSW_RX_VLAN_ENCAP_HDR_PRIO_MSK GENMASK(2, 0)
@@ -454,6 +463,9 @@ struct cpsw_priv {
u8 mac_addr[ETH_ALEN];
bool rx_pause;
bool tx_pause;
+ bool mqprio_hw;
+ int fifo_bw[CPSW_TC_NUM];
+ int shp_cfg_speed;
u32 emac_port;
struct cpsw_common *cpsw;
};
@@ -968,8 +980,8 @@ static int cpsw_tx_mq_poll(struct napi_struct *napi_tx, int budget)
/* process every unprocessed channel */
ch_map = cpdma_ctrl_txchs_state(cpsw->dma);
- for (ch = 0, num_tx = 0; ch_map; ch_map >>= 1, ch++) {
- if (!(ch_map & 0x01))
+ for (ch = 0, num_tx = 0; ch_map & 0xff; ch_map <<= 1, ch++) {
+ if (!(ch_map & 0x80))
continue;
txv = &cpsw->txv[ch];
@@ -1078,6 +1090,38 @@ static void cpsw_set_slave_mac(struct cpsw_slave *slave,
slave_write(slave, mac_lo(priv->mac_addr), SA_LO);
}
+static bool cpsw_shp_is_off(struct cpsw_priv *priv)
+{
+ struct cpsw_common *cpsw = priv->cpsw;
+ struct cpsw_slave *slave;
+ u32 shift, mask, val;
+
+ val = readl_relaxed(&cpsw->regs->ptype);
+
+ slave = &cpsw->slaves[cpsw_slave_index(cpsw, priv)];
+ shift = CPSW_FIFO_SHAPE_EN_SHIFT + 3 * slave->slave_num;
+ mask = 7 << shift;
+ val = val & mask;
+
+ return !val;
+}
+
+static void cpsw_fifo_shp_on(struct cpsw_priv *priv, int fifo, int on)
+{
+ struct cpsw_common *cpsw = priv->cpsw;
+ struct cpsw_slave *slave;
+ u32 shift, mask, val;
+
+ val = readl_relaxed(&cpsw->regs->ptype);
+
+ slave = &cpsw->slaves[cpsw_slave_index(cpsw, priv)];
+ shift = CPSW_FIFO_SHAPE_EN_SHIFT + 3 * slave->slave_num;
+ mask = (1 << --fifo) << shift;
+ val = on ? val | mask : val & ~mask;
+
+ writel_relaxed(val, &cpsw->regs->ptype);
+}
+
static void _cpsw_adjust_link(struct cpsw_slave *slave,
struct cpsw_priv *priv, bool *link)
{
@@ -1117,6 +1161,12 @@ static void _cpsw_adjust_link(struct cpsw_slave *slave,
mac_control |= BIT(4);
*link = true;
+
+ if (priv->shp_cfg_speed &&
+ priv->shp_cfg_speed != slave->phy->speed &&
+ !cpsw_shp_is_off(priv))
+ dev_warn(priv->dev,
+ "Speed was changed, CBS shaper speeds are changed!");
} else {
mac_control = 0;
/* disable forwarding */
@@ -1578,6 +1628,231 @@ static void cpsw_slave_stop(struct cpsw_slave *slave, struct cpsw_common *cpsw)
soft_reset_slave(slave);
}
+static int cpsw_tc_to_fifo(int tc, int num_tc)
+{
+ if (tc == num_tc - 1)
+ return 0;
+
+ return CPSW_FIFO_SHAPERS_NUM - tc;
+}
+
+static int cpsw_set_fifo_bw(struct cpsw_priv *priv, int fifo, int bw)
+{
+ struct cpsw_common *cpsw = priv->cpsw;
+ u32 val = 0, send_pct, shift;
+ struct cpsw_slave *slave;
+ int pct = 0, i;
+
+ if (bw > priv->shp_cfg_speed * 1000)
+ goto err;
+
+ /* shaping has to stay enabled for highest fifos linearly
+ * and fifo bw no more then interface can allow
+ */
+ slave = &cpsw->slaves[cpsw_slave_index(cpsw, priv)];
+ send_pct = slave_read(slave, SEND_PERCENT);
+ for (i = CPSW_FIFO_SHAPERS_NUM; i > 0; i--) {
+ if (!bw) {
+ if (i >= fifo || !priv->fifo_bw[i])
+ continue;
+
+ dev_warn(priv->dev, "Prev FIFO%d is shaped", i);
+ continue;
+ }
+
+ if (!priv->fifo_bw[i] && i > fifo) {
+ dev_err(priv->dev, "Upper FIFO%d is not shaped", i);
+ return -EINVAL;
+ }
+
+ shift = (i - 1) * 8;
+ if (i == fifo) {
+ send_pct &= ~(CPSW_PCT_MASK << shift);
+ val = DIV_ROUND_UP(bw, priv->shp_cfg_speed * 10);
+ if (!val)
+ val = 1;
+
+ send_pct |= val << shift;
+ pct += val;
+ continue;
+ }
+
+ if (priv->fifo_bw[i])
+ pct += (send_pct >> shift) & CPSW_PCT_MASK;
+ }
+
+ if (pct >= 100)
+ goto err;
+
+ slave_write(slave, send_pct, SEND_PERCENT);
+ priv->fifo_bw[fifo] = bw;
+
+ dev_warn(priv->dev, "set FIFO%d bw = %d\n", fifo,
+ DIV_ROUND_CLOSEST(val * priv->shp_cfg_speed, 100));
+
+ return 0;
+err:
+ dev_err(priv->dev, "Bandwidth doesn't fit in tc configuration");
+ return -EINVAL;
+}
+
+static int cpsw_set_fifo_rlimit(struct cpsw_priv *priv, int fifo, int bw)
+{
+ struct cpsw_common *cpsw = priv->cpsw;
+ struct cpsw_slave *slave;
+ u32 tx_in_ctl_rg, val;
+ int ret;
+
+ ret = cpsw_set_fifo_bw(priv, fifo, bw);
+ if (ret)
+ return ret;
+
+ slave = &cpsw->slaves[cpsw_slave_index(cpsw, priv)];
+ tx_in_ctl_rg = cpsw->version == CPSW_VERSION_1 ?
+ CPSW1_TX_IN_CTL : CPSW2_TX_IN_CTL;
+
+ if (!bw)
+ cpsw_fifo_shp_on(priv, fifo, bw);
+
+ val = slave_read(slave, tx_in_ctl_rg);
+ if (cpsw_shp_is_off(priv)) {
+ /* disable FIFOs rate limited queues */
+ val &= ~(0xf << CPSW_FIFO_RATE_EN_SHIFT);
+
+ /* set type of FIFO queues to normal priority mode */
+ val &= ~(3 << CPSW_FIFO_QUEUE_TYPE_SHIFT);
+
+ /* set type of FIFO queues to be rate limited */
+ if (bw)
+ val |= 2 << CPSW_FIFO_QUEUE_TYPE_SHIFT;
+ else
+ priv->shp_cfg_speed = 0;
+ }
+
+ /* toggle a FIFO rate limited queue */
+ if (bw)
+ val |= BIT(fifo + CPSW_FIFO_RATE_EN_SHIFT);
+ else
+ val &= ~BIT(fifo + CPSW_FIFO_RATE_EN_SHIFT);
+ slave_write(slave, val, tx_in_ctl_rg);
+
+ /* FIFO transmit shape enable */
+ cpsw_fifo_shp_on(priv, fifo, bw);
+ return 0;
+}
+
+/* Defaults:
+ * class A - prio 3
+ * class B - prio 2
+ * shaping for class A should be set first
+ */
+static int cpsw_set_cbs(struct net_device *ndev,
+ struct tc_cbs_qopt_offload *qopt)
+{
+ struct cpsw_priv *priv = netdev_priv(ndev);
+ struct cpsw_common *cpsw = priv->cpsw;
+ struct cpsw_slave *slave;
+ int prev_speed = 0;
+ int tc, ret, fifo;
+ u32 bw = 0;
+
+ tc = netdev_txq_to_tc(priv->ndev, qopt->queue);
+
+ /* enable channels in backward order, as highest FIFOs must be rate
+ * limited first and for compliance with CPDMA rate limited channels
+ * that also used in bacward order. FIFO0 cannot be rate limited.
+ */
+ fifo = cpsw_tc_to_fifo(tc, ndev->num_tc);
+ if (!fifo) {
+ dev_err(priv->dev, "Last tc%d can't be rate limited", tc);
+ return -EINVAL;
+ }
+
+ /* do nothing, it's disabled anyway */
+ if (!qopt->enable && !priv->fifo_bw[fifo])
+ return 0;
+
+ /* shapers can be set if link speed is known */
+ slave = &cpsw->slaves[cpsw_slave_index(cpsw, priv)];
+ if (slave->phy && slave->phy->link) {
+ if (priv->shp_cfg_speed &&
+ priv->shp_cfg_speed != slave->phy->speed)
+ prev_speed = priv->shp_cfg_speed;
+
+ priv->shp_cfg_speed = slave->phy->speed;
+ }
+
+ if (!priv->shp_cfg_speed) {
+ dev_err(priv->dev, "Link speed is not known");
+ return -1;
+ }
+
+ ret = pm_runtime_get_sync(cpsw->dev);
+ if (ret < 0) {
+ pm_runtime_put_noidle(cpsw->dev);
+ return ret;
+ }
+
+ bw = qopt->enable ? qopt->idleslope : 0;
+ ret = cpsw_set_fifo_rlimit(priv, fifo, bw);
+ if (ret) {
+ priv->shp_cfg_speed = prev_speed;
+ prev_speed = 0;
+ }
+
+ if (bw && prev_speed)
+ dev_warn(priv->dev,
+ "Speed was changed, CBS shaper speeds are changed!");
+
+ pm_runtime_put_sync(cpsw->dev);
+ return ret;
+}
+
+static void cpsw_cbs_resume(struct cpsw_slave *slave, struct cpsw_priv *priv)
+{
+ int fifo, bw;
+
+ for (fifo = CPSW_FIFO_SHAPERS_NUM; fifo > 0; fifo--) {
+ bw = priv->fifo_bw[fifo];
+ if (!bw)
+ continue;
+
+ cpsw_set_fifo_rlimit(priv, fifo, bw);
+ }
+}
+
+static void cpsw_mqprio_resume(struct cpsw_slave *slave, struct cpsw_priv *priv)
+{
+ struct cpsw_common *cpsw = priv->cpsw;
+ u32 tx_prio_map = 0;
+ int i, tc, fifo;
+ u32 tx_prio_rg;
+
+ if (!priv->mqprio_hw)
+ return;
+
+ for (i = 0; i < 8; i++) {
+ tc = netdev_get_prio_tc_map(priv->ndev, i);
+ fifo = CPSW_FIFO_SHAPERS_NUM - tc;
+ tx_prio_map |= fifo << (4 * i);
+ }
+
+ tx_prio_rg = cpsw->version == CPSW_VERSION_1 ?
+ CPSW1_TX_PRI_MAP : CPSW2_TX_PRI_MAP;
+
+ slave_write(slave, tx_prio_map, tx_prio_rg);
+}
+
+/* restore resources after port reset */
+static void cpsw_restore(struct cpsw_priv *priv)
+{
+ /* restore MQPRIO offload */
+ for_each_slave(priv, cpsw_mqprio_resume, priv);
+
+ /* restore CBS offload */
+ for_each_slave(priv, cpsw_cbs_resume, priv);
+}
+
static int cpsw_ndo_open(struct net_device *ndev)
{
struct cpsw_priv *priv = netdev_priv(ndev);
@@ -1657,6 +1932,8 @@ static int cpsw_ndo_open(struct net_device *ndev)
}
+ cpsw_restore(priv);
+
/* Enable Interrupt pacing if configured */
if (cpsw->coal_intvl != 0) {
struct ethtool_coalesce coal;
@@ -2191,6 +2468,78 @@ static int cpsw_ndo_set_tx_maxrate(struct net_device *ndev, int queue, u32 rate)
return ret;
}
+static int cpsw_set_mqprio(struct net_device *ndev, void *type_data)
+{
+ struct tc_mqprio_qopt_offload *mqprio = type_data;
+ struct cpsw_priv *priv = netdev_priv(ndev);
+ struct cpsw_common *cpsw = priv->cpsw;
+ int fifo, num_tc, count, offset;
+ struct cpsw_slave *slave;
+ u32 tx_prio_map = 0;
+ int i, tc, ret;
+
+ num_tc = mqprio->qopt.num_tc;
+ if (num_tc > CPSW_TC_NUM)
+ return -EINVAL;
+
+ if (mqprio->mode != TC_MQPRIO_MODE_DCB)
+ return -EINVAL;
+
+ ret = pm_runtime_get_sync(cpsw->dev);
+ if (ret < 0) {
+ pm_runtime_put_noidle(cpsw->dev);
+ return ret;
+ }
+
+ if (num_tc) {
+ for (i = 0; i < 8; i++) {
+ tc = mqprio->qopt.prio_tc_map[i];
+ fifo = cpsw_tc_to_fifo(tc, num_tc);
+ tx_prio_map |= fifo << (4 * i);
+ }
+
+ netdev_set_num_tc(ndev, num_tc);
+ for (i = 0; i < num_tc; i++) {
+ count = mqprio->qopt.count[i];
+ offset = mqprio->qopt.offset[i];
+ netdev_set_tc_queue(ndev, i, count, offset);
+ }
+ }
+
+ if (!mqprio->qopt.hw) {
+ /* restore default configuration */
+ netdev_reset_tc(ndev);
+ tx_prio_map = TX_PRIORITY_MAPPING;
+ }
+
+ priv->mqprio_hw = mqprio->qopt.hw;
+
+ offset = cpsw->version == CPSW_VERSION_1 ?
+ CPSW1_TX_PRI_MAP : CPSW2_TX_PRI_MAP;
+
+ slave = &cpsw->slaves[cpsw_slave_index(cpsw, priv)];
+ slave_write(slave, tx_prio_map, offset);
+
+ pm_runtime_put_sync(cpsw->dev);
+
+ return 0;
+}
+
+static int cpsw_ndo_setup_tc(struct net_device *ndev, enum tc_setup_type type,
+ void *type_data)
+{
+ switch (type) {
+ case TC_SETUP_QDISC_CBS:
+ return cpsw_set_cbs(ndev, type_data);
+
+ case TC_SETUP_QDISC_MQPRIO:
+ return cpsw_set_mqprio(ndev, type_data);
+
+ default:
+ return -EOPNOTSUPP;
+ }
+}
+
static const struct net_device_ops cpsw_netdev_ops = {
.ndo_open = cpsw_ndo_open,
.ndo_stop = cpsw_ndo_stop,
@@ -2206,6 +2555,7 @@ static const struct net_device_ops cpsw_netdev_ops = {
#endif
.ndo_vlan_rx_add_vid = cpsw_ndo_vlan_rx_add_vid,
.ndo_vlan_rx_kill_vid = cpsw_ndo_vlan_rx_kill_vid,
+ .ndo_setup_tc = cpsw_ndo_setup_tc,
};
static int cpsw_get_regs_len(struct net_device *ndev)
@@ -2432,7 +2782,7 @@ static int cpsw_update_channels_res(struct cpsw_priv *priv, int ch_num, int rx)
void (*handler)(void *, int, int);
struct netdev_queue *queue;
struct cpsw_vector *vec;
- int ret, *ch;
+ int ret, *ch, vch;
if (rx) {
ch = &cpsw->rx_ch_num;
@@ -2445,7 +2795,8 @@ static int cpsw_update_channels_res(struct cpsw_priv *priv, int ch_num, int rx)
}
while (*ch < ch_num) {
- vec[*ch].ch = cpdma_chan_create(cpsw->dma, *ch, handler, rx);
+ vch = rx ? *ch : 7 - *ch;
+ vec[*ch].ch = cpdma_chan_create(cpsw->dma, vch, handler, rx);
queue = netdev_get_tx_queue(priv->ndev, *ch);
queue->tx_maxrate = 0;
@@ -2982,7 +3333,7 @@ static int cpsw_probe(struct platform_device *pdev)
u32 slave_offset, sliver_offset, slave_size;
const struct soc_device_attribute *soc;
struct cpsw_common *cpsw;
- int ret = 0, i;
+ int ret = 0, i, ch;
int irq;
cpsw = devm_kzalloc(&pdev->dev, sizeof(struct cpsw_common), GFP_KERNEL);
@@ -3157,7 +3508,8 @@ static int cpsw_probe(struct platform_device *pdev)
if (soc)
cpsw->quirk_irq = 1;
- cpsw->txv[0].ch = cpdma_chan_create(cpsw->dma, 0, cpsw_tx_handler, 0);
+ ch = cpsw->quirk_irq ? 0 : 7;
+ cpsw->txv[0].ch = cpdma_chan_create(cpsw->dma, ch, cpsw_tx_handler, 0);
if (IS_ERR(cpsw->txv[0].ch)) {
dev_err(priv->dev, "error initializing tx dma channel\n");
ret = PTR_ERR(cpsw->txv[0].ch);
diff --git a/drivers/net/ethernet/ti/davinci_cpdma.c b/drivers/net/ethernet/ti/davinci_cpdma.c
index 4f1267477aa4b56b7f3e1d19420302728da56e7d..4236dcdd56348512d6d2393ea27c19c0bd9078bf 100644
--- a/drivers/net/ethernet/ti/davinci_cpdma.c
+++ b/drivers/net/ethernet/ti/davinci_cpdma.c
@@ -406,37 +406,36 @@ static int cpdma_chan_fit_rate(struct cpdma_chan *ch, u32 rate,
struct cpdma_chan *chan;
u32 old_rate = ch->rate;
u32 new_rmask = 0;
- int rlim = 1;
+ int rlim = 0;
int i;
- *prio_mode = 0;
for (i = tx_chan_num(0); i < tx_chan_num(CPDMA_MAX_CHANNELS); i++) {
chan = ctlr->channels[i];
- if (!chan) {
- rlim = 0;
+ if (!chan)
continue;
- }
if (chan == ch)
chan->rate = rate;
if (chan->rate) {
- if (rlim) {
- new_rmask |= chan->mask;
- } else {
- ch->rate = old_rate;
- dev_err(ctlr->dev, "Prev channel of %dch is not rate limited\n",
- chan->chan_num);
- return -EINVAL;
- }
- } else {
- *prio_mode = 1;
- rlim = 0;
+ rlim = 1;
+ new_rmask |= chan->mask;
+ continue;
}
+
+ if (rlim)
+ goto err;
}
*rmask = new_rmask;
+ *prio_mode = rlim;
return 0;
+
+err:
+ ch->rate = old_rate;
+ dev_err(ctlr->dev, "Upper cpdma ch%d is not rate limited\n",
+ chan->chan_num);
+ return -EINVAL;
}
static u32 cpdma_chan_set_factors(struct cpdma_ctlr *ctlr,