ConnectX-3 devices can use either 64- or 32-byte completion queue
entries (CQEs) and event queue entries (EQEs). Using 64-byte
EQEs/CQEs performs better because each entry is aligned to a complete
cacheline. This patch queries the HCA's capabilities, and if it
supports 64-byte CQEs and EQES the driver will configure the HW to
work in 64-byte mode.
The 32-byte vs 64-byte mode is global per HCA and not per CQ or EQ.
Since this mode is global, userspace (libmlx4) must be updated to work
with the configured CQE size, and guests using SR-IOV virtual
functions need to know both EQE and CQE size.
In case one of the 64-byte CQE/EQE capabilities is activated, the
patch makes sure that older guest drivers that use the QUERY_DEV_FUNC
command (e.g as done in mlx4_core of Linux 3.3..3.6) will notice that
they need an update to be able to work with the PPF. This is done by
changing the returned pf_context_behaviour not to be zero any more. In
case none of these capabilities is activated that value remains zero
and older guest drivers can run OK.
The SRIOV related flow is as follows
1. the PPF does the detection of the new capabilities using
QUERY_DEV_CAP command.
2. the PPF activates the new capabilities using INIT_HCA.
3. the VF detects if the PPF activated the capabilities using
QUERY_HCA, and if this is the case activates them for itself too.
Note that the VF detects that it must be aware to the new PF behaviour
using QUERY_FUNC_CAP. Steps 1 and 2 apply also for native mode.
User space notification is done through a new field introduced in
struct mlx4_ib_ucontext which holds device capabilities for which user
space must take action. This changes the binary interface so the ABI
towards libmlx4 exposed through uverbs is bumped from 3 to 4 but only
when **needed** i.e. only when the driver does use 64-byte CQEs or
future device capabilities which must be in sync by user space. This
practice allows to work with unmodified libmlx4 on older devices (e.g
A0, B0) which don't support 64-byte CQEs.
In order to keep existing systems functional when they update to a
newer kernel that contains these changes in VF and userspace ABI, a
module parameter enable_64b_cqe_eqe must be set to enable 64-byte
mode; the default is currently false.
Signed-off-by: Eli Cohen <eli@mellanox.com>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: Roland Dreier <roland@purestorage.com>
On an SR-IOV master device, __mlx4_init_one() calls mlx4_init_hca()
before mlx4_multi_func_init(). However, for unlucky configurations,
mlx4_init_hca() might call mlx4_SENSE_PORT() (via mlx4_dev_cap()), and
that calls mlx4_cmd_imm() with MLX4_CMD_WRAPPED set.
However, on a multifunction device with MLX4_CMD_WRAPPED, __mlx4_cmd()
calls into mlx4_slave_cmd(), and that immediately tries to do
down(&priv->cmd.slave_sem);
but priv->cmd.slave_sem isn't initialized until mlx4_multi_func_init()
(which we haven't called yet). The next thing it tries to do is access
priv->mfunc.vhcr, but that hasn't been allocated yet.
Fix this by moving the initialization of slave_sem and vhcr up into
mlx4_cmd_init(). Also, since slave_sem is really just being used as a
mutex, convert it into a slave_cmd_mutex.
Signed-off-by: Roland Dreier <roland@purestorage.com>
This is necessary in order to support > 1 VF/PF in a VM for software
that uses the node guid as a discriminator, such as librdmacm.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <roland@purestorage.com>
1. Allow only master to change node description.
2. Prevent AH leakage in send mads.
3. Take device part number from PCI structure, so that guests see the
VF part number (and not the PF part number).
4. Place the device revision ID into caps structure at startup.
5. SET_PORT in update_gids_task needs to go through wrapper on master.
6. In mlx4_ib_event(), PORT_MGMT_EVENT needs be handled in a work
queue on the master, since it propagates events to slaves using
GEN_EQE.
7. Do not support FMR on slaves.
8. Add spinlock to slave_event(), since it is called both in interrupt
context and in process context (due to 6 above, and also if
smp_snoop is used). This fix was found and implemented by Saeed
Mahameed <saeedm@mellanox.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <roland@purestorage.com>
For IB ports, we paravirtualize the GUID at index 0 on slaves. The
GUID at index 0 seen by a slave is the actual GUID occupying the GUID
table at the slave-id index.
The driver, by default, requests at startup time that subnet manager
populate its entire guid table with GUIDs. These guids are then mapped
(paravirtualized) to the slaves, and appear for each slave as its GUID
at index 0.
Until each slave has such a guid, its port status is DOWN.
The guid table is cached to support special QP paravirtualization, and
event propagation to slaves on guid change (we test to see if the guid
really changed before propagating an event to the slave).
To support this caching, add capability to __mlx4_ib_query_gid() to
obtain the network view (i.e., physical view) gid at index X, not just
the host (paravirtualized) view.
Based on a patch from Erez Shitrit <erezsh@mellanox.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <roland@purestorage.com>
The MAD_IFC firmware command fulfills two functions.
First, it is used in the QP0/QP1 MAD-handling flow to obtain
information from the FW (for answering queries), and for setting
variables in the HCA (MAD SET packets).
For this, MAD_IFC should provide the FW (physical) view of the data.
This is the view that OpenSM needs. We call this the "network view".
In the second case, MAD_IFC is used by various verbs to obtain data
regarding the local HCA (e.g., ib_query_device()). We call this the
"host view".
This data needs to be paravirtualized.
MAD_IFC therefore needs a wrapper function, and also needs another
flag indicating whether it should provide the network view (when it is
called by ib_process_mad in special-qp packet handling), or the host
view (when it is called while implementing a verb).
There are currently 2 flag parameters in mlx4_MAD_IFC already:
ignore_bkey and ignore_mkey. These two parameters are replaced by a
single "mad_ifc_flags" parameter, with different bits set for each
flag. A third flag is added: "network-view/host-view".
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <roland@purestorage.com>
This requires:
1. Replacing the paravirtualized P_Key index (inserted by the guest)
with the real P_Key index.
2. For UD QPs, placing the guest's true source GID index in the
address path structure mgid field, and setting the ud_force_mgid
bit so that the mgid is taken from the QP context and not from the
WQE when posting sends.
3. For UC and RC QPs, placing the guest's true source GID index in the
address path structure mgid field.
4. For tunnel and proxy QPs, setting the Q_Key value reserved for that
proxy/tunnel pair.
Since not all the above adjustments occur in all the QP transitions,
the QP transitions require separate wrapper functions.
Secondly, initialize the P_Key virtualization table to its default
values: Master virtualized table is 1-1 with the real P_Key table,
guest virtualized table has P_Key index 0 mapped to the real P_Key
index 0, and all the other P_Key indices mapped to the reserved
(invalid) P_Key at index 127.
Finally, add logic in smp_snoop for maintaining the phys_P_Key_cache.
and generating events on the master only if a P_Key actually changed.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Allocate SR-IOV paravirtualization resources and MAD demuxing contexts
on the master.
This has two parts. The first part is to initialize the structures to
contain the contexts. This is done at master startup time in
mlx4_ib_init_sriov().
The second part is to actually create the tunneling resources required
on the master to support a slave. This is performed the master
detects that a slave has started up (MLX4_DEV_EVENT_SLAVE_INIT event
generated when a slave initializes its comm channel).
For the master, there is no such startup event, so it creates its own
tunneling resources when it starts up. In addition, the master also
creates the real special QPs. The ib_core layer on the master causes
creation of proxy special QPs, since the master is also
paravirtualized at the ib_core layer.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Roland Dreier <roland@purestorage.com>
Currently the mlx4 drivers don't have the necessary callbacks to
implement EEH errors detection and recovery, so the PCI layer uses the
probe and remove callbacks to try to recover the device after an error on
the bus. However, these callbacks have race conditions with the internal
catastrophic error recovery functions, which will also detect the error
and this can cause the system to crash if both EEH and catas functions
try to reset the device.
This patch adds the necessary error recovery callbacks and makes sure
that the internal catastrophic error functions will not try to reset the
device in such scenarios. It also adds some calls to
pci_channel_offline() to suppress reads/writes on the bus when the slot
cannot accept I/O operations so we prevent unnecessary accesses to the
bus and speed up the device removal.
Signed-off-by: Kleber Sacilotto de Souza <klebers@linux.vnet.ibm.com>
Acked-by: Shlomo Pongratz <shlomop@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Add support for firmware commands to attach/detach a new device managed
steering mode. Such network steering rules allow the user to provide an
L2/L3/L4 flow specification to the firmware and have the device to steer
traffic that matches that specification to the provided QP.
Signed-off-by: Hadar Hen Zion <hadarh@mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
- pass the following parameters:
- firmware version (added QUERY_FW paravirtualization for that)
- disable Blueflame on slaves. KVM disables write combining on guests,
and we get better performance without BF in this case. (This requires
QUERY_DEV_CAP paravirtualization, also in this commit)
- max qp rdma as destination
- get rid of a chunk of "if (0)" dead code
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Reviewed-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Currently the slave and master resources are deleted after master freed
all bitmaps. If any resources were not properly cleaned up during the
shutdown process, an Oops would result.
Fix so that delete slave (only) resources during cleanup. Master resources
are cleaned up during unload process, and need not separately be cleaned.
Note that during cleanup, we need to split the resource-tracker freeing
functionality.
Before removing all the bitmaps, we free any leftover slave resources.
However, we can only remove the resource tracker linked list after
all bitmap frees, since some of the freeing functions (e.g.,
mlx4_cleanup_eq_table) use paravirtualized FW commands which expect
the resource tracker linked list to be present.
Found-by: Aviad Yehezkel <aviadye@mellanox.com>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Handle the compiler warnings on variables which are set but not used
by removing the relevant variable or casting a return value which is
ignored on purpose to void.
Signed-off-by: Or Gerlitz <ogerlitz@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Prevent race condition between commands on comm channel.
Happened while unloading the driver when switching from
event to polling mode. VF got completion on the last command
before switching to polling mode, but toggle was not changed.
After the fix - VF will not write the next command before
toggle is updated.
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
Perform cleanup also in non-master flow.
The VFs use communication channel as well.
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
Correct spelling "reseting" to "resetting" in
drivers/net/ethernet/mellanox/mlx4/cmd.c
Signed-off-by: Masanari Iida <standby24x7@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Slave async events were mapped to single eq. This patch fixes this issue, so
the slaves can map the async events to any eq.
Signed-off-by: Marcel Apfelbaum <marcela@dev.mellanox.co.il>
Reviewed-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
The communication channel is HW interface from PF point of view
So the command return status should be stored as HW error code
and only then translated to errno values.
Reporetd-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
1. Added module parameters sr_iov and probe_vf for controlling enablement of
SRIOV mode.
2. Increased default max num-qps, num-mpts and log_num_macs to accomodate
SRIOV mode
3. Added port_type_array as a module parameter to allow driver startup with
ports configured as desired.
In SRIOV mode, only ETH is supported, and this array is ignored; otherwise,
for the case where the FW supports both port types (ETH and IB), the
port_type_array parameter is used.
By default, the port_type_array is set to configure both ports as IB.
4. When running in sriov mode, the master needs to initialize the ICM eq table
to hold the eq's for itself and also for all the slaves.
5. mlx4_set_port_mask() now invoked from mlx4_init_hca, instead of in mlx4_dev_cap.
6. Introduced sriov VF (slave) device startup/teardown logic (mainly procedures
mlx4_init_slave, mlx4_slave_exit, mlx4_slave_cap, mlx4_slave_exit and flow
modifications in __mlx4_init_one, mlx4_init_hca, and mlx4_setup_hca).
VFs obtain their startup information from the PF (master) device via the
comm channel.
7. In SRIOV mode (both PF and VF), MSI_X must be enabled, or the driver
aborts loading the device.
8. Do not allow setting port type via sysfs when running in SRIOV mode.
9. mlx4_get_ownership: Currently, only one PF is supported by the driver.
If the HCA is burned with FW which enables more than one PF, only one
of the PFs is allowed to run. The first one up grabs a FW ownership
semaphone -- all other PFs will find that semaphore taken, and the
driver will not allow them to run.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: Liran Liss <liranl@mellanox.co.il>
Signed-off-by: Marcel Apfelbaum <marcela@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
The physical port is now common to the PF and VFs.
The port resources and configuration is managed by the PF, VFs can
only influence the MTU of the port, it is set as max among all functions,
Each function allocates RX buffers of required size to meet it's MTU enforcement.
Port management code was moved to mlx4_core, as the mlx4_en module is
virtualization unaware
Move handling qp functionality to mlx4_get_eth_qp/mlx4_put_eth_qp
including reserve/release range and add/release unicast steering.
Let mlx4_register/unregister_mac deal only with MAC (un)registration.
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
Let multicast/unicast attaching flow go through resource tracker.
The PF is the one responsible for managing all the steering entries.
Define and use module parameter that determines the number of qps
per multicast group.
Minor changes in function calls according to changed prototype.
Signed-off-by: Eugenia Emantayev <eugenia@mellanox.co.il>
Signed-off-by: Yevgeny Petrilin <yevgenyp@mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
The resource tracker is used to track usage of HCA resources by the different
guests.
Virtual functions (VFs) are attached to guest operating systems but
resources are allocated from the same pool and are assigned to VFs. It is
essential that hostile/buggy guests not be able to affect the operation of
other VFs, possibly attached to other guest OSs since ConnectX firmware is not
tolerant to misuse of resources.
The resource tracker module associates each resource with a VF and maintains
state information for the allocated object. It also defines allowed state
transitions and enforces them.
Relationships between resources are also referred to. For example, CQs are
pointed to by QPs, so it is forbidden to destroy a CQ if a QP refers to it.
ICM memory is always accessible through the primary function and hence it is
allocated by the owner of the primary function.
When a guest dies, an FLR is generated for all the VFs it owns and all the
resources it used are freed.
The tracked resource types are: QPs, CQs, SRQs, MPTs, MTTs, MACs, RES_EQs,
and XRCDNs.
Signed-off-by: Eli Cohen <eli@mellanox.co.il>
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
When SRIOV is enabled, pf and vfs communicate via shared comm channel.
The vf gets its side of the comm channel via a VF BAR.
Each VF (slave) creates its vHCR (virtual HCA Command Register),
Its DMA address is passed to the PF (master) using Communication Channel Register.
The same Register is used to notify the master of commands posted by the
slaves and for the master to pass events to the slaves, such as command completions
and asynchronous events.
The vHCR format is identical to the HCR format, except for the 'go' and 't' bits,
which are reserved in the vHCR. Posting commands to the vHCR is identical to
the way it is done with the HCR, albeit that the function/PF token fields are
used instead of the HCR go bit.
Specifically:
- When the function prepares a new command in the vHCR, it issues the Post_vHCR_cmd
communication channel command and toggles the value of the function token;
when PF token has an equal value, the command has been accepted and a new command may be posted.
- When the PF detects a Post_vHCR_cmd command, it concludes that a new command is available in the vHCR;
after processing the command, the PF toggles the PF token to match the function token.
When the 'e' bit is not set, the completion of a Post_vHCR_cmd command also indicates
the completion the vHCR command. If, however, the 'e' bit is set, the completion of a
Post_vHCR_cmd command only indicates that the vHCR command has been accepted for execution by the PF.
Function commands are processed by the PF as follows:
-DMA (using the ACCESS_MEM command) the vHCR image into a shadow buffer.
-Validate that the opcode is non-privileged, and that the opcode- and input-modifiers are legal.
-DMA the in-box (if required) into a shadow buffer.
-Validate the command:
o Resource ranges (e.g., QP ranges).
o Partition key.
o Ranges of referenced resources (e.g., CQs within QP contexts).
-If the 'e' bit is set
o complete the Post_vHCR_cmd command
-Execute the command on the HCR.
-DMA the results to the vHCR out-box (if required).
-If the 'e' bit is set
o Indicate command completion by generating a completion event using the GEN_EQE command
-Otherwise
o DMA the command status to the vHCR
o Complete the Post_vHCR_cmd command
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: Yevgeny Petrillin <yevgenyp@mellanox.com>
Signed-off-by: Liran Liss <liranl@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
For SRIOV, some Hypervisor commands can be executed directly (native = 1).
Others should go through the command wrapper flow (for tracking resource
usage, for example, or for changing some HCA configurations that slaves
need to be notified of).
This patch sets the groundwork for this capability -- adding the correct
value of "native" in each case.
Note that if SRIOV is not activated, this parameter has no effect.
Signed-off-by: Jack Morgenstein <jackm@dev.mellanox.co.il>
Signed-off-by: David S. Miller <davem@davemloft.net>
These were getting the macros from an implicit module.h
include via device.h, but we are planning to clean that up.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
drivers/net: Add export.h to wireless/brcm80211/brcmfmac/bcmsdh.c
This relatively recently added file uses EXPORT_SYMBOL and hence
needs export.h included so that it is compatible with the module.h
split up work.
Signed-off-by: Paul Gortmaker <paul.gortmaker@windriver.com>
Moves the Mellanox driver into drivers/net/ethernet/mellanox/ and
make the necessary Kconfig and Makefile changes.
CC: Roland Dreier <roland@kernel.org>
Signed-off-by: Jeff Kirsher <jeffrey.t.kirsher@intel.com>
Or Gerlitz