Signed-off-by: Marty Connor <mdc@etherboot.org>

Enable interrupts in sis900_irq(). Doing so allows some programs using
gPXE's UNDI interface to work properly, including Symantec Ghost.

Tested-by: Hubert Mercier <hubert.mercier@unilim.fr>
Signed-off-by: Marty Connor <mdc@etherboot.org>

Both methods disabled packet tx and rx just to have it enabled again
by calling a3c90x_reset().
Fixed by disabling tx and rx after the call to a3c90x_reset().

Tested by booting Ubuntu intrepid(8.10) directly from gPXE and pxelinux.
Tested on 3c905, 3c905B, 3c905C.

Signed-off-by: Marty Connor <mdc@etherboot.org>

Some systems will retry their boot sequence in the event of a boot
failure.  On these systems, the second and subsequent boot attempts
will fail to initialise the Hermon HCA.

Fix by resetting the HCA during probe().  This incurs a one-second
cost, but there seems to be no viable alternative.

Originally-fixed-by: Itay Gazit <itaygazit@gmail.com>

Some devices can only be reset via a mechanism that also resets the
card's PCI core, thus necessitating a backup and restore of all or
part of the PCI configuration space across a reset.

Signed-off-by: Marty Connor <mdc@etherboot.org>

IPoIB has a 20-byte link-layer address, of which only eight bytes
represent anything relating to a "hardware address".

The PXE and EFI SNP APIs expect the permanent address to be the same
size as the link-layer address, so fill in the "permanent address"
field with the initial link layer address (as generated by
register_netdev() based upon the real hardware address).

The hardware address is an intrinsic property of the hardware, while
the link-layer address can be changed at runtime.  This separation is
exposed via APIs such as PXE and EFI, but is currently elided by gPXE.

Expose the hardware and link-layer addresses as separate properties
within a net device.  Drivers should now fill in hw_addr, which will
be used to initialise ll_addr at the time of calling
register_netdev().

With iSCSI, connection attempts are expensive; it may take many
seconds to determine that a connection will fail.  SRP connection
attempts are much less expensive, so we may as well avoid the
"optimisation" of declaring a state of permanent failure after a
certain number of attempts.  This allows a gPXE SRP initiator to
resume operations after an arbitrary amount of SRP target downtime.

SRP is the SCSI RDMA Protocol.  It allows for a method of SAN booting
whereby the target is responsible for reading and writing data using
Remote DMA directly to the initiator's memory.  The software initiator
merely sends and receives SCSI commands; it never has to touch the
actual data.

The ACK timeout determines how long we take to notice a failed
Reliable Connection.  Reducing it from the arbitrary value of 19 down
to 14 reduces the individual ACK timeout from around 2.1s to 67ms;
this in turn reduces the time to tear down and re-establish a broken
SRP session from around 30s to around 1s.

The Infiniband Communication Manager will refuse to establish a
connection if it believes the connection is already established.
There is no immediately obvious way to ask it to tear down the
existing connection and replace it; to issue a DREP we would need to
know the local and remote communication IDs used for the previous
connection setup.

We can work around this by randomising the high-order bits of the
queue pair number; these have no significance to the hardware, but are
sufficient to convince the IB CM that this is a different connection.

Signed-off-by: Michael Brown <mcb30@etherboot.org>

The prior net80211 model of physical-layer behavior for drivers was
overly simplistic and limited the drivers that could be written.  To
be more flexible, split the driver-provided list of supported rates by
band, and add a means for specifying a list of supported channels.
Allow drivers to specify a hardware channel value that will be tied to
uses of the channel.

Expose net80211_duration() to drivers, and make the rate it uses in
its computations configurable, so that it can be used in calculating
durations that must be set in hardware for ACK and CTS packets. Add
net80211_cts_duration() for the common case of calculating the
duration for a CTS packet.

Signed-off-by: Michael Brown <mcb30@etherboot.org>

Originally-fixed-by: Itay Gazit <itaygazit@gmail.com>

The IBA specification refers to management "interfaces" and "agents".
The interface is the component that connects to the queue pair and
sends and receives MADs; the agent is the component that constructs
the reply to the MAD.

Rename the IB_{QPN,QKEY,QPT} constants as a first step towards making
this separation in gPXE.

Signed-off-by: Michael Brown <mcb30@etherboot.org>

The Linux IB Communication Manager will always send MADs to QP1,
rather than back to the originating QP.  On Hermon, QP1 is by default
handled by the embedded firmware.  We can change this, but the cost is
that we have to handle both QP0 and QP1 (i.e. we have to provide SMA
as well as GMA service in software), and we have to use MLX queues
rather than standard UD queues (i.e. we have to construct the UD
datagrams by hand).

There doesn't seem to be any viable way around this situation, ugly
though it is.

Queue pairs are now assumed to be created in the INIT state, with a
call to ib_modify_qp() required to bring the queue pair to the RTS
state.

ib_modify_qp() no longer takes a modification list; callers should
modify the relevant queue pair parameters (e.g. qkey) directly and
then call ib_modify_qp() to synchronise the changes to the hardware.

The packet sequence number is now a property of the queue pair, rather
than of the device.

Each queue pair may have an associated address vector.  For RC queue
pairs, this is the address vector that will be programmed in to the
hardware as the remote address.  For UD queue pairs, it will be used
as the default address vector if none is supplied to ib_post_send().

Attempting the broadcast group join while the link is down is
harmless, but can generate annoying volumes of debug messages.

The GMA code was based upon the SMA code.  We can save space by making
the SMA simply an instance of the GMA.

Now that IPoIB has to deal with only one set of queues, the queue set
abstraction becomes merely an inconvenient wrapper.

All packets handled by the metadata queue set now go via the GMA.

Generalise out the multicast group membership record code from IPoIB.

The queue key is stored as a property of the queue pair, and so can
optionally be added by the Infiniband core at the time of calling
ib_post_send(), rather than always having to be specified by the
caller.

This allows IPoIB to avoid explicitly keeping track of the data queue
key.

Now that path record lookups are handled entirely via
ib_resolve_path(), the only role of the IPoIB peer cache is as a
lookup table for MAC addresses.  Update the code structure and
comments to reflect this.

The IPoIB broadcast MAC address varies according to the partition key.
Now that the broadcast MAC address is a property of the network device
rather than of the link layer, we can expose this real MAC address
directly.

The broadcast LID is now identified via a path record lookup; this is
marginally inefficient (since it was present in the MCMemberRecord
GetResponse), but avoids the need to special-case broadcasts when
constructing the address vector in ipoib_transmit().

Remove the special handling of the IPoIB broadcast QPN.

Generalise out the path record lookup code from IPoIB.

Currently, all Infiniband users must create a process for polling
their completion queues (or rely on a regular hook such as
netdev_poll() in ipoib.c).

Move instead to a model whereby the Infiniband core maintains a single
process calling ib_poll_eq(), and polling the event queue triggers
polls of the applicable completion queues.  (At present, the
Infiniband core simply polls all of the device's completion queues.)
Polling a completion queue will now implicitly refill all attached
receive work queues; this is analogous to the way that netdev_poll()
implicitly refills the RX ring.

Infiniband users no longer need to create a process just to poll their
completion queues and refill their receive rings.

IPoIB and the SMA have separate constants for the packet size to be
used to I/O buffer allocations.  Merge these into the single
IB_MAX_PAYLOAD_SIZE constant.

(Various other points in the Infiniband stack have hard-coded
assumptions of a 2048-byte payload; we don't currently support
variable MTUs.)

IPoIB has a link-layer broadcast address that varies according to the
partition key.  We currently go through several contortions to pretend
that the link-layer address is a fixed constant; by making the
broadcast address a property of the network device rather than the
link-layer protocol it will be possible to simplify IPoIB's broadcast
handling.

Move the icky call to step() from aoe.c to ata.c; this takes it at
least one step further away from where it really doesn't belong.

Unfortunately, AoE has the ugly aoe_discover() mechanism which means
that we still have a step() loop in aoe.c for now; this needs to be
replaced at some future point.