linux.git/drivers/net/dsa, branch v4.2-rc4

net: dsa: bcm_sf2: do not use indirect reads and writes for 7445E0

2015-07-20T23:47:30+00:00

7445E0 contains an ECO which disconnected the internal SF2 pseudo-PHY which was
known to conflict with the external pseudo-PHY of BCM53125 switches. This
motivated the need to utilize the internal SF2 MDIO controller via indirect
register reads/writes to control external Broadcom switches due to this address
conflict (both responded at address 30d).

For 7445E0, the internal pseudo-PHY of the SF2 switch got disconnected, and as
a consequence this prevents the internal SF2 MDIO bus controller from reading
data (reads back everything as 0) since the MDI line is tied low.

Fix this by making the indirect register reads and writes conditional to
7445D0, on 7445E0 we can utilize the SWITCH_MDIO controller (backed by
mdio-unimac and not the DSA created slave MII bus).

We utilize of_machine_is_compatible() here since this is the only way for use
to differentiate between these two chips in a way that does not violate layers
or becomes (too) vendor-specific.

Signed-off-by: Florian Fainelli 
Signed-off-by: David S. Miller

net: dsa: mv88e6xxx: fix fid_mask when leaving bridge

2015-07-20T19:44:14+00:00

The mv88e6xxx_priv_state structure contains an fid_mask, where 1 means
the FID is free to use, 0 means the FID is in use.

This patch fixes the bit clear in mv88e6xxx_leave_bridge() when
assigning a new FID to a port.

Example scenario: I have 7 ports, port 5 is CPU, port 6 is unused (no
PHY). After setting the ports 0, 1 and 2 in bridge br0, and ports 3 and
4 in bridge br1, I have the following fid_mask: 0b111110010110 (0xf96).

Indeed, br0 uses FID 0, and br1 uses FID 3.

After setting nomaster for port 0, I get the wrong fid_mask: 0b10 (0x2).

With this patch we correctly get 0b111110010100 (0xf94), meaning port 0
uses FID 1, br0 uses FID 0, and br1 uses FID 3.

Signed-off-by: Vivien Didelot 
Reviewed-by: Guenter Roeck 
Signed-off-by: David S. Miller

dsa: mv88x6xxx: Zero statistics counters

2015-06-23T13:37:17+00:00

Zero the statistics counters when setting up the global
registers. Otherwise the counters will remain from the last boot if
the power has not been removed.

Signed-off-by: Andrew Lunn 
Signed-off-by: David S. Miller

dsa: mv88x6xxx: Add debugfs interface for scratch registers

2015-06-23T13:33:42+00:00

Allow the contents of the scratch registers to be shown in debugfs.

Signed-off-by: Andrew Lunn 
Signed-off-by: David S. Miller

dsa: mv88x6xxx: Add debugfs interface for device map

2015-06-23T13:33:41+00:00

The device map is used to route packets between cascaded switches.
Add dumping a switches device map via debugfs.

Signed-off-by: Andrew Lunn 
Signed-off-by: David S. Miller

dsa: mv88x6xxx: Add debugfs interface for statistics

2015-06-23T13:33:41+00:00

Allow the contents of the statistics counters to be shown in debugfs.
This is particularly useful for the cpu and dsa ports, which cannot be
seen using ethtools -S.

Signed-off-by: Andrew Lunn 
Signed-off-by: David S. Miller

dsa: mv88x6xxx: Refactor getting a single statistic

2015-06-23T13:33:40+00:00

Move the code to retrieve a statistics counter into a function of its
own, so it can later be reused.

Signed-off-by: Andrew Lunn 
Signed-off-by: David S. Miller

dsa: mv88e6xxx: Add debugfs interface for ATU

2015-06-23T13:33:40+00:00

Dump the Address Translation Unit via a file in debugfs.

Signed-off-by: Andrew Lunn 
Signed-off-by: David S. Miller

dsa: mv88e6xxx: Add debugfs interface for registers

2015-06-23T13:33:39+00:00

Allow the contents of the registers to be shown in debugfs.

Signed-off-by: Andrew Lunn 
Signed-off-by: David S. Miller

net: dsa: bcm_sf2: Utilize BRCM_PSEUDO_PHY_ADDR

2015-06-11T06:33:58+00:00

Utilize the newly introduced BRCM_PSEUDO_PHY_ADDR constant from
brcmphy.h instead of open-coding the Broadcom Ethernet switches
pseudo-PHY address (30).

Signed-off-by: Florian Fainelli 
Signed-off-by: David S. Miller