We didn't specify values for MaxRecvDataSegmentLength and
MaxBurstLength (to save space, since we were happy with the
RFC-defined default values of 8kB and 256kB respectively).  However,
the OpenSolaris target (incorrectly) assumes default values of zero
for these parameters.

The upshot was that the OpenSolaris target would get stuck in an
endless loop trying to send us the first 512-byte sector, zero bytes
at a time, and would eventually run out of memory and core-dump.

Fixed by explicitly specifying the default values for these two
parameters.

packet, to work around the problem whereby small TX rings get backed
up because we haven't yet poll()ed for TX completions.

packet, to work around the problem whereby small TX rings get backed
up because we haven't yet poll()ed for TX completions.

to construct DHCP packets suitable for Infiniband.

tracking down a bug that turned out to be a free_iob() used where I
needed a netdev_tx_complete().  This left the freed I/O buffer on the
net device's TX list, with bad, bad consequences later.

Also fixed the bug in question.

makes it easier to put the generic allocation code into infiniband.c

using entirely our own code.

fixed; still using a quick hack-up just to be able to pass through
data.

to construct DHCP packets suitable for Infiniband.

(Problem observed by Clay McClure in VMware Fusion.)

order in which they occur.

to generate the initial transmission; this cuts off around 0.3s per
instantiated connection.

list.

Added FEATURE() macros to most relevant (non-driver) files.

and generate DHCP options to indicate features to DHCP server (and to
PXE NBPs).

algorithms; we only need the pubkey check disabled (and only because
pubkey algorithms are not yet integrated into the crypto_algorithm
subsystem).

iSCSI hack.

really ugly hack at present, but that doesn't hugely matter since I'm
aiming to change the interface to iSCSI devices anyway within the next
week.

in tcp_rx_data().

Clarify comments on discarding duplicate or out-of-order data.