Add PRM structures to support Hermon Ethernet devices.
Signed-off-by: Itay Gazit <itaygazit@gmail.com>
Modified-by: Michael Brown <mcb30@ipxe.org>
Signed-off-by: Michael Brown <mcb30@ipxe.org>
Unlike Arbel, port parameters must be applied via a separate call to
SET_PORT, rather than as parameters to INIT_PORT.
Signed-off-by: Michael Brown <mcb30@ipxe.org>
Mapping a single page at a time causes a several-second delay at
device initialisation time. Reduce this by mapping multiple pages at
a time, using the largest block sizes possible given the alignment
constraints.
Signed-off-by: Michael Brown <mcb30@ipxe.org>
Only port state change events are currently mapped to our event queue,
since those are the only events we are prepared to handle. This
ignores a potentially useful source of diagnostic information in the
case of unexpected failures.
Fix by mapping all events to the event queue; a build with debugging
enabled will therefore at least dump the raw content of the unexpected
events.
Signed-off-by: Michael Brown <mcb30@ipxe.org>
No event is generated upon reaching INIT, so we must poll separately
for link state changes while we remain DOWN.
Signed-off-by: Michael Brown <mcb30@ipxe.org>
ib_smc_update() potentially updates the Infiniband port state, and so
should almost always be followed by a call to ib_link_state_changed().
The one exception is the call made to ib_smc_update() before the
device is registered.
Fix by removing explicit calls to ib_link_state_changed() from drivers
using ib_smc_update(), including a call to ib_link_state_changed()
within ib_smc_update(), and creating a separate ib_smc_init() for use
prior to device registration.
Signed-off-by: Michael Brown <mcb30@ipxe.org>
Doorbell records are currently embedded within the completion queue
and receive work queue strucures, which are allocated using zalloc()
and so have an alignment guarantee of only sizeof(void*), i.e. four
bytes. This is sufficient for the receive work queue, but not for the
completion queue, which requires an alignment guarantee of eight
bytes.
Though not guaranteed, it so happens that zalloc() will always return
a pointer that is exactly four bytes above a sixteen-byte boundary.
The completion queue doorbell record is therefore always misaligned,
and the value passed to the hardware via SW2HW_CQ is actually always
pointing to the page_offset value within the MTT descriptor (which
directly precedes the inline doorbell record). Provided that the page
offset is greater than 0x100, this looks to the hardware like an
update_ci value of greater than 0x010000 (taking into account
endianness differences), and so the hardware will happily deliver more
than 0x010000 completions before stopping. Hence this problem is
rarely observable.
Fix by allocating the doorbell records separately and using the
correct alignment constraints.
Signed-off-by: Michael Brown <mcb30@ipxe.org>
Give completion queues a chance to deliver exception events by
programming in the number of our event queue (currently used only for
port state changes).
Signed-off-by: Michael Brown <mcb30@ipxe.org>
Improve the utility of debugging messages by including the relevant
port number, queue number (QPN, CQN, EQN), work queue entry (WQE)
number, and physical addresses wherever applicable.
Add hermon_dump_cqctx() for dumping a completion queue context, and
hermon_fill_nop_send_wqe() for inserting NOPs into send work queues.
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>
Align each ICM member alloc to the member size instead of page size.
Increase multicast table size to 128.
Signed-off-by: Itay Gazit <itaygazit@gmail.com>
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>
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.
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.
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().
The patch file supplied for commit 3a799e9 ("[hermon] Add PCI ID for
ConnectX QDR card") accidentally marked drivers/infiniband/hermon.c as
being executable.
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.
Not all Infiniband cards have embedded subnet management agents.
Split out the code that communicates with such an embedded SMA into a
separate ib_smc.c file, and have drivers call ib_smc_update()
explicitly when they suspect that the answers given by the embedded
SMA may have changed.
Receive completion handlers now get passed an address vector
containing the information extracted from the packet headers
(including the GRH, if present), and only the payload remains in the
I/O buffer.
This breaks the symmetry between transmit and receive completions, so
remove the ib_completer_t type and use an ib_completion_queue_operations
structure instead.
Rename the "destination QPN" and "destination LID" fields in struct
ib_address_vector to reflect its new dual usage.
Since the ib_completion structure now contains only an IB status code,
("syndrome") replace it with a generic gPXE integer status code.
Avoid leaking I/O buffers in ib_destroy_qp() by completing any
outstanding work queue entries with a generic error code. This
requires the completion handlers to be available to ib_destroy_qp(),
which is done by making them static configuration parameters of the CQ
(set by ib_create_cq()) rather than being provided on each call to
ib_poll_cq().
This mimics the functionality of netdev_{tx,rx}_flush(). The netdev
flush functions would previously have been catching any I/O buffers
leaked by the IPoIB data queue (though not by the IPoIB metadata
queue).
Use individual page mappings rather than a single whole-region
mapping, to avoid the waste of memory that occurs due to the
constraint that each mapped block must be aligned on its own size.
We were accidentally allocating only half the required amount of
memory (given the alignment method) for the firmware buffer, leading
to conflicts between the firmware buffer and gPXE code/data segments.
Infiniband devices no longer block waiting for link-up in
register_ibdev().
Hermon driver needs to create an event queue and poll for link-up events.
Infiniband core needs to reread MAD parameters when link state changes.
IPoIB needs to cope with Infiniband link parameters being only partially
available at probe and open time.
Arbel and Hermon cards both have multiple ports. Add the
infrastructure required to register each port as a separate IB
device. Don't yet register more than one port, since registration
will currently fail unless a valid link is detected.
Use ib_*_{set,get}_{drv,owner}data wrappers to access driver- and
owner-private data on Infiniband structures.
Pull out common code for handling management datagrams from arbel.c
and hermon.c into infiniband.c.
Add port number to struct ib_device.
Add open(), close() and mad() methods to struct ib_device_operations.
