linux.git/drivers/net/ethernet/sfc, branch v3.7

ptp: use list_move instead of list_del/list_add

2012-10-07T18:52:14+00:00

Using list_move() instead of list_del() + list_add().

dpatch engine is used to auto generate this patch.
(https://github.com/weiyj/dpatch)

Signed-off-by: Wei Yongjun 
Signed-off-by: David S. Miller

drivers/net/ethernet/sfc: use standard __{clear,set}_bit_le() functions

2012-10-05T18:04:54+00:00

There are now standard functions for dealing with little-endian bit
arrays, so use them instead of our own implementations.

Signed-off-by: Ben Hutchings 
Signed-off-by: Takuya Yoshikawa 
Cc: David Miller 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

Merge branch 'for-davem' of git://git.kernel.org/pub/scm/linux/kernel/git/bwh/sfc-next

2012-10-02T01:32:55+00:00

Ben Hutchings says:

====================
Some bug fixes that should go into 3.7:

1. Fix oops when removing device with SR-IOV enabled.  (This regression
was introduced by the last set of changes, so the fix does not need to
be applied to any earlier kernel versions.)
2. Fix firmware structure field lookup bug that resulted in missing
sensor information.
3. Fix bug that makes self-test do very little in some configurations.
4. Fix the numbering of ethtool RX flow steering filters to reflect the
real hardware priorities.
====================

Signed-off-by: David S. Miller

sfc: Fix the reported priorities of different filter types

2012-10-02T00:58:42+00:00

Each RX filter table contains filters with two different levels of
specificity: TCP/IPv4 and UDP/IPv4 filters match the local address and
port and optionally the remote address and port; Ethernet filters
match the local address and optionally the VID.  The more specific
filters always override less specific filters within the same table,
and should be numbered accordingly.

Signed-off-by: Ben Hutchings

sfc: Remove EFX_FILTER_FLAG_RX_OVERRIDE_IP

2012-10-02T00:58:41+00:00

This filter flag cannot yet be set through the ethtool command and
will not be supported on future hardware.

Signed-off-by: Ben Hutchings

sfc: Fix loopback self-test with separate_tx_channels=1

2012-10-02T00:58:41+00:00

The loopback self-test iterates over all the TX queues of channel 0,
which is not very interesting when that's an RX-only channel.

Signed-off-by: Ben Hutchings

sfc: Fix MCDI structure field lookup

2012-10-02T00:58:40+00:00

The least significant bit number (LBN) of a field within an MCDI
structure is counted from the start of the structure, not the
containing dword.  In MCDI_ARRAY_FIELD() we need to mask it rather
than using the usual EFX_DWORD_FIELD() macro.

Signed-off-by: Ben Hutchings

sfc: Add parentheses around use of bitfield macro arguments

2012-10-02T00:58:39+00:00

Signed-off-by: Ben Hutchings

sfc: Fix null function pointer in efx_sriov_channel_type

2012-10-02T00:57:25+00:00

Commit c31e5f9 ('sfc: Add channel specific receive_skb handler and
post_remove callback') added the function pointer field
efx_channel_type::post_remove and an unconditional call through it.

This field should have been initialised to efx_channel_dummy_op_void
in the existing instances of efx_channel_type, but this was only done
in efx_default_channel_type.  Consequently, if a device has SR-IOV
enabled then removing the driver or device will result in an oops.

Signed-off-by: Ben Hutchings

ptp: link the phc device to its parent device

2012-09-22T19:42:38+00:00

PTP Hardware Clock devices appear as class devices in sysfs. This patch
changes the registration API to use the parent device, clarifying the
clock's relationship to the underlying device.

Signed-off-by: Richard Cochran 
Acked-by: Ben Hutchings 
Acked-by: Jeff Kirsher 
Signed-off-by: David S. Miller