On POWER systems, DMA mapping/unmapping operations are very expensive.
These changes reduce these costs by trying to reuse DMA mapped pages.

After all the buffers associated with a page have been processed and
passed up, the page is placed into a ring (if there is room).  For
each page that is required for a refill operation, a page in the ring
is examined to determine if its page count has fallen to 1, ie. the
kernel has released its reference to these packets.  If this is the
case, the page can be immediately added back into the RX descriptor
ring, without having to re-map it for DMA.

If the kernel is still holding a reference to this page, it is removed
from the ring and unmapped for DMA.  Then a new page, which can
immediately be used by RX buffers in the descriptor ring, is allocated
and DMA mapped.

The time a page needs to spend in the recycle ring before the kernel
has released its page references is based on the number of buffers
that use this page.  As large pages can hold more RX buffers, the RX
recycle ring can be shorter.  This reduces memory usage on POWER
systems, while maintaining the performance gain achieved by recycling
pages, following the driver change to pack more than two RX buffers
into large pages.

When an IOMMU is not present, the recycle ring can be small to reduce
memory usage, since DMA mapping operations are inexpensive.

With a small recycle ring, attempting to refill the descriptor queue
with more buffers than the equivalent size of the recycle ring could
ultimately lead to memory leaks if page entries in the recycle ring
were overwritten.  To prevent this, the check to see if the recycle
ring is full is changed to check if the next entry to be written is
NULL.

[bwh: Combine and rebase several commits so this is complete
 before the following buffer-packing changes.  Remove module
 parameter.]
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

Enable RX DMA scattering iff an RX buffer large enough for the current
MTU will not fit into a single page and the NIC supports DMA
scattering for kernel-mode RX queues.

On Falcon and Siena, the RX_USR_BUF_SIZE field is used as the DMA
limit for both all RX queues with scatter enabled.  Set it to 1824,
matching what Onload uses now.

Maintain a statistic for frames truncated due to lack of descriptors
(rx_nodesc_trunc).  This is distinct from rx_frm_trunc which may be
incremented when scattering is disabled and implies an over-length
frame.

Whenever an MTU change causes scattering to be turned on or off,
update filters that point to the PF queues, but leave others
unchanged, as VF drivers assume scattering is off.

Add n_frags parameters to various functions, and make them iterate:
- efx_rx_packet()
- efx_recycle_rx_buffers()
- efx_rx_mk_skb()
- efx_rx_deliver()

Make efx_handle_rx_event() responsible for updating
efx_rx_queue::removed_count.

Change the RX pipeline state to a starting ring index and number of
fragments, and make __efx_rx_packet() responsible for clearing it.

Based on earlier versions by David Riddoch and Jon Cooper.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

The pipeline mechanism will need to change a bit for scattered
packets.  Add a wrapper to insulate efx_process_channel() from this.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

Replace efx_nic::rx_buffer_len with efx_nic::rx_dma_len, the maximum
RX DMA length.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

The Linux side of EEH is triggered by MMIO reads, but this
driver's data path does not issue any MMIO reads (except in
legacy interrupt mode).  Therefore add a monitor function
to poll EEH periodically.

When preparing to reset the device based on our own error
detection, also poll EEH and defer to its recovery mechanism
if appropriate.

[bwh: Use a separate condition for the initial link poll; fix some
 style errors]
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

[bwh: Remove more dead code, and make efx_ptp_rx() pull the data it
 needs into the header area.]
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

We must only ever stop TX queues when they are full or the net device
is not 'ready' so far as the net core, and specifically the watchdog,
is concerned.  Otherwise, the watchdog may fire *immediately* if no
packets have been added to the queue in the last 5 seconds.

The device is ready if all the following are true:

(a) It has a qdisc
(b) It is marked present
(c) It is running
(d) The link is reported up

(a) and (c) are normally true, and must not be changed by a driver.
(d) is under our control, but fake link changes may disturb userland.
This leaves (b).  We already mark the device absent during reset
and self-test, but we need to do the same during MTU changes and ring
reallocation.  We don't need to do this when the device is brought
down because then (c) is already false.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

The __dev* removal patches for the network drivers ended up messing up
the function prototypes for a bunch of drivers.  This patch fixes all of
them back up to be properly aligned.

Bonus is that this almost removes 100 lines of code, always a nice
surprise.
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

CONFIG_HOTPLUG is going away as an option.  As result the __dev*
markings will be going away.

Remove use of __devinit, __devexit_p, __devinitdata, __devinitconst,
and __devexit.
Signed-off-by: NBill Pemberton <wfp5p@virginia.edu>
Cc: Solarflare linux maintainers <linux-net-drivers@solarflare.com>
Cc: Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@linuxfoundation.org>

Most of the module parameters treated as boolean are currently exposed
as type int or uint.  Defining them with the proper type is useful
documentation for both users and developers.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

The assertion of netif_device_present() at the top of
efx_hard_start_xmit() may fail if we don't do this.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

There are now standard functions for dealing with little-endian bit
arrays, so use them instead of our own implementations.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>
Signed-off-by: NTakuya Yoshikawa <yoshikawa.takuya@oss.ntt.co.jp>
Cc: David Miller <davem@davemloft.net>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Add PTP IEEE-1588 support and make accesible via the PHC subsystem.

This work is based on prior code by Andrew Jackson
Signed-off-by: NStuart Hodgson <smhodgson@solarflare.com>
[bwh:
 - Add byte order conversion in efx_ptp_send_times()
 - Simplify conversion of PPS event times
 - Add the built-in vs module check to CONFIG_SFC_PTP dependencies]
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

Allows an extra channel to override the standard receive_skb handler
and also for extra non generic operations to be performed on remove.

Also set default rx strategy so only skbs can be delivered to the
PTP receive function.
Signed-off-by: NStuart Hodgson <smhodgson@solarflare.com>
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

The PTP channel will have its own RX queue even though it's not
a regular traffic channel.

Original work by Ben Hutchings <bhutchings@solarflare.com>
Signed-off-by: NStuart Hodgson <smhodgson@solarflare.com>
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

Following commit 8f4cccbb ('net: Set device operstate at registration
time') it is now correct and preferable to set the carrier off before
registering a device.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

We also stop clearing *efx in efx_init_struct().  This is safe because
alloc_etherdev_mq() already clears it for us.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

RX DMA is limited by the length specified in each descriptor and not
by the MAC.  Over-length frames may get into the RX FIFO regardless of
the MAC settings, due to a hardware bug, but they will be truncated by
the packet DMA engine and reported as such in the completion event.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

We try to defer resets while the device is not READY, but we're not
doing this quite correctly.  In particular, changes to efx_nic::state
are documented as serialised by the RTNL lock, but they aren't.

1. We check whether a reset was requested during probe (suggesting
broken hardware) before we allow requested resets to be scheduled.
This leaves a window where a requested reset would be deferred
indefinitely.

2. Although we cancel the reset work item during device removal,
there are still later operations that can cause it to be scheduled
again.  We need to check the state before scheduling it.

3. Since the state can change between scheduling and running of
the work item, we still need to check it there, and we need to
do so *after* acquiring the RTNL lock which serialises state
changes.

4. We must cancel the reset work item during device removal, if the
state could ever have been READY.  This wasn't done in some of the
failure paths from efx_pci_probe().  Move the cancellation to
efx_pci_remove_main().
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

The current informational message doesn't properly explain what
happens, and could also appear if we defer a reset during
suspend/resume.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

efx_change_mtu() and efx_realloc_channels() each stop and start much
of the NIC, even if it has been disabled.  Since efx_start_all() is a
no-op when the NIC is disabled, this is probably harmless in the case
of efx_change_mtu(), but efx_realloc_channels() also reenables
interrupts which could be a bad thing to do.

Change efx_start_all() and efx_start_interrupts() to assert that the
NIC is not disabled, but make efx_stop_interrupts() do nothing if the
NIC is disabled (since it is already stopped), consistent with
efx_stop_all().

Update comments for efx_start_all() and efx_stop_all() to describe
their purpose and preconditions more accurately.

Add a common function to check and log if the NIC is disabled, and use
it in efx_net_open(), efx_change_mtu() and efx_realloc_channels().
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

Interrupt state should be consistently guarded by the RTNL lock once
the net device is registered.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

Currently we ignore and clear the disabled state.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

I don't think these PM functions can race with userland net device
operations, but it's much easier to reason about locking if state is
consistently guarded by the same lock.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

STATE_INIT and STATE_FINI are equivalent and represent incompletely
initialised states; combine them as STATE_UNINIT.

Rename STATE_RUNNING to STATE_READY, to avoid confusion with
netif_running() and IFF_RUNNING.

The comments do not quite match current usage, but this will be
corrected in subsequent fixes.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

We now have a definite upper bound on the number of descriptors per
skb; use that to stop the queue when the next packet might not fit.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

Currently an skb requiring TSO may not fit within a minimum-size TX
queue.  The TX queue selected for the skb may stall and trigger the TX
watchdog repeatedly (since the problem skb will be retried after the
TX reset).  This issue is designated as CVE-2012-3412.

Set the maximum number of TSO segments for our devices to 100.  This
should make no difference to behaviour unless the actual MSS is less
than about 700.  Increase the minimum TX queue size accordingly to
allow for 2 worst-case skbs, so that there will definitely be space
to add an skb after we wake a queue.

To avoid invalidating existing configurations, change
efx_ethtool_set_ringparam() to fix up values that are too small rather
than returning -EINVAL.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

We need to clear the private data pointer in the PCI device.
Also reorder cleanup in efx_pci_remove() for symmetry.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

efx_nic_fatal_interrupt() disables DMA before scheduling a reset.
After this, we need not and *cannot* flush queues.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

If RSS is disabled on the PF (efx->n_rx_channels == 1) we try to set
up the indirection table so that VFs can use it, setting
efx->rss_spread = efx_vf_size(efx).  But if SR-IOV was disabled at
compile time, this evaluates to 0 and we end up dividing by zero when
initialising the table.

I considered changing the fallback definition of efx_vf_size() to
return 1, but its value is really meaningless if we are not going to
enable VFs.  Therefore add a condition of efx_sriov_wanted(efx) in
efx_probe_interrupts().
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

During probe of each port, read and log the part number from VPD.
Remove the Falcon-specific board name lookup.

Initial version by Stuart Hodgson <smhodgson@solarflare.com>.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

Generate a test event on each event queue whenever the interface is
brought up, then after 1 second check that we have either handled a
test event or handled another IRQ for each event queue.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

RX and TX completions on the same event queue are generally not associated
with the same flows.  The inclusion of TX completions in the adaptive IRQ
score is more of a source of noise rather than useful feedback.  Therefore,
do not include them in the score, and adjust the default threshold scores
down.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

Unify return value of .ndo_set_mac_address if the given address
isn't valid. Return -EADDRNOTAVAIL as eth_mac_addr() already does
if is_valid_ether_addr() fails.
Signed-off-by: NDanny Kukawka <danny.kukawka@bisect.de>
Signed-off-by: NDavid S. Miller <davem@davemloft.net>

We have a very simple way of allocating buffer table entries to
queues, which is just to take the next one available.  The extra
channels are the highest numbered channels but they need to be
allocated the lowest entries so that the traffic channels can be
allocated new entries without any collisions.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

On the SFC9000 family, each port has 1024 Virtual Interfaces (VIs),
each with an RX queue, a TX queue, an event queue and a mailbox
register.  These may be assigned to up to 127 SR-IOV virtual functions
per port, with up to 64 VIs per VF.

We allocate an extra channel (IRQ and event queue only) to receive
requests from VF drivers.

There is a per-port limit of 4 concurrent RX queue flushes, and queue
flushes may be initiated by the MC in response to a Function Level
Reset (FLR) of a VF.  Therefore, when SR-IOV is in use, we submit all
flush requests via the MC.

The RSS indirection table is shared with VFs, so the number of RX
queues used in the PF is limited to the number of VIs per VF.

This is almost entirely the work of Steve Hodgson, formerly
shodgson@solarflare.com.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

Each port has a block of 64-bit SRAM that is divided between buffer
table and descriptor cache regions at initialisation time.  Currently
we use a fixed allocation, but it needs to be changed to support
larger numbers of queues.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>

This lets us identify the NIC affected in case of failure, and
will be necessary to adjust for SR-IOV constraints.
Signed-off-by: NBen Hutchings <bhutchings@solarflare.com>