The Infiniband link status change callback ipoib_link_state_changed()
may be called while the IPoIB device is closed, in which case there
will not be an IPoIB queue pair to be joined to the IPv4 broadcast
group. This leads to NULL pointer dereferences in ib_mcast_attach()
and ib_mcast_detach().
Fix by not attempting to join (or leave) the broadcast group unless we
actually have an IPoIB queue pair.
Signed-off-by: Wissam Shoukair <wissams@mellanox.com>
Modified-by: Michael Brown <mcb30@ipxe.org>
Signed-off-by: Michael Brown <mcb30@ipxe.org>
Multicast MAC addresses will never have REMAC cache entries, and the
corresponding multicast IPoIB MAC address cannot be obtained simply by
issuing an ARP request.
For the trivial volume of multicast packets that we expect to send in
any realistic scenario, the simplest solution is to send them as
broadcasts instead.
Reported-by: Wissam Shoukair <wissams@mellanox.com>
Signed-off-by: Michael Brown <mcb30@ipxe.org>
The only way to map an eIPoIB MAC address (REMAC) to an IPoIB MAC
address is to intercept an incoming ARP request or reply.
If we do not have an REMAC cache entry for a particular destination
MAC address, then we cannot transmit the packet. This can arise in at
least two situations:
- An external program (e.g. a PXE NBP using the UNDI API) may attempt
to transmit to a destination MAC address that has been obtained by
some method other than ARP.
- Memory pressure may have caused REMAC cache entries to be
discarded. This is fairly likely on a busy network, since REMAC
cache entries are created for all received (broadcast) ARP
requests. (We can't sensibly avoid creating these cache entries,
since they are required in order to send an ARP reply, and when we
are being used via the UNDI API we may have no knowledge of which
IP addresses are "ours".)
Attempt to ameliorate the situation by generating a semi-spurious ARP
request whenever we find a missing REMAC cache entry. This will
hopefully trigger an ARP reply, which would then provide us with the
information required to populate the REMAC cache.
Signed-off-by: Michael Brown <mcb30@ipxe.org>
As with the neighbour cache, discarding an REMAC cache entry is
potentially very disruptive.
Originally-fixed-by: Wissam Shoukair <wissams@mellanox.com>
Signed-off-by: Michael Brown <mcb30@ipxe.org>
Almost all clients of the raw-packet interfaces (UNDI and SNP) can
handle only Ethernet link layers. Expose an Ethernet-compatible link
layer to local clients, while remaining compatible with IPoIB on the
wire. This requires manipulation of ARP (but not DHCP) packets within
the IPoIB driver.
This is ugly, but it's the only viable way to allow IPoIB devices to
be driven via the raw-packet interfaces.
Signed-off-by: Michael Brown <mcb30@ipxe.org>
Avoid spurious matches for peer key 0 against empty peer cache
entries, and set the LL_MULTICAST flag in addition to LL_BROADCAST.
Signed-off-by: Michael Brown <mcb30@ipxe.org>
Allow the link layer to directly report whether or not a packet is
multicast or broadcast at the time of calling pull(), rather than
relying on heuristics to determine this at a later stage.
Signed-off-by: Michael Brown <mcb30@ipxe.org>
Replace the explicit calls from the Infiniband core to the IPoIB layer
with the general concept of an Infiniband upper-layer driver
(analogous to a PCI driver) which can create arbitrary devices on top
of Infiniband devices.
Signed-off-by: Michael Brown <mcb30@ipxe.org>
Access to the gpxe.org and etherboot.org domains and associated
resources has been revoked by the registrant of the domain. Work
around this problem by renaming project from gPXE to iPXE, and
updating URLs to match.
Also update README, LOG and COPYRIGHTS to remove obsolete information.
Signed-off-by: Michael Brown <mcb30@ipxe.org>
The first byte of the IPoIB MAC address is used for flags indicating
support for "connected mode". Strip out the non-QPN bits of the first
dword when constructing the address vector for transmitted IPoIB
packets, so as not to end up passing an invalid QPN in the BTH.
The iBFT is Ethernet-centric in providing only six bytes for a MAC
address. This is most probably an indirect consequence of a similar
design flaw in the Windows NDIS stack. (The WinOF IPoIB stack
performs all sorts of contortions in order to pretend to the NDIS
layer that it is dealing with six-byte MAC addresses.)
There is no sensible way in which to extend the iBFT without breaking
compatibility with programs that expect to parse it. Add the notion
of an "Ethernet-compatible" MAC address to our link layer abstraction,
so that link layers can provide their own workarounds for this
limitation.
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().
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().
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().
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.
In order to construct outgoing link-layer frames or parse incoming
ones properly, some protocols (such as 802.11) need more state than is
available in the existing variables passed to the link-layer protocol
handlers. To remedy this, add struct net_device *netdev as the first
argument to each of these functions, so that more information can be
fetched from the link layer-private part of the network device.
Updated all three call sites (netdevice.c, efi_snp.c, pxe_undi.c) and
both implementations (ethernet.c, ipoib.c) of ll_protocol to use the
new argument.
Signed-off-by: Michael Brown <mcb30@etherboot.org>
This brings us in to line with Linux definitions, and also simplifies
adding x86_64 support since both platforms have 2-byte shorts, 4-byte
ints and 8-byte long longs.
The return path in directed route SMPs lists the egress ports in order
from SM to node, rather than from node to SM.
To write to the correct offset within the return path, we need to
parse the hop pointer. This is held within the class-specific data
portion of the MAD header, which was previously unused by us and
defined to be a uint16_t. Define this field to be a union type; this
requires some rearrangement of ib_mad.h and corresponding changes to
ipoib.c.
Michael Brown
Wissam Shoukair
Michael Brown
Joshua Oreman