link_down is self-explanatory.
rdma_sends and rdma_recvs count the number of RDMA Send and RDMA Receive
operations completed successfully. This is different from the existing
sent_pkts and rcvd_pkts counters because the existing counters measure
packets, not RDMA operations.
ack_deffered is renamed to ack_deferred to fix the spelling.
out_of_sequence is renamed to out_of_seq_request to make clear that it is
counting only requests and not other packets which can be out of sequence.
Signed-off-by: Andrew Boyer <andrew.boyer@dell.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
In ib_query_port(), use the netdev's IFF_UP flag to determine phys_state
(flag set = down = POLLING, flag clear = disabled = DISABLED).
Callers can then use the phys_state field to distinguish between links
which have a dead partner, cable missing, etc., from links which are
turned off on the local node. This is useful for HA and supportability.
Signed-off-by: Andrew Boyer <andrew.boyer@dell.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Error completions must still contain a valid wr_id and
qp_num such that the consumer can rely on. Correctly
fill these fields in receive error completions.
Reported-by: Walker Benjamin <benjamin.walker@intel.com>
Cc: stable@vger.kernel.org
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Tested-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
When resp is in error state, the queued SKBs will not be handled.
The function get_req cleans up the skb queue directly.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Use rdma_set_device_sysfs_group() to register device attributes and
simplify the driver.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Clang warns when an emumerated type is implicitly converted to another.
drivers/infiniband/sw/rxe/rxe.c:106:27: warning: implicit conversion
from enumeration type 'enum rxe_device_param' to different enumeration
type 'enum ib_atomic_cap' [-Wenum-conversion]
rxe->attr.atomic_cap = RXE_ATOMIC_CAP;
~ ^~~~~~~~~~~~~~
drivers/infiniband/sw/rxe/rxe.c:131:22: warning: implicit conversion
from enumeration type 'enum rxe_port_param' to different enumeration
type 'enum ib_port_state' [-Wenum-conversion]
port->attr.state = RXE_PORT_STATE;
~ ^~~~~~~~~~~~~~
drivers/infiniband/sw/rxe/rxe.c:132:24: warning: implicit conversion
from enumeration type 'enum rxe_port_param' to different enumeration
type 'enum ib_mtu' [-Wenum-conversion]
port->attr.max_mtu = RXE_PORT_MAX_MTU;
~ ^~~~~~~~~~~~~~~~
drivers/infiniband/sw/rxe/rxe.c:133:27: warning: implicit conversion
from enumeration type 'enum rxe_port_param' to different enumeration
type 'enum ib_mtu' [-Wenum-conversion]
port->attr.active_mtu = RXE_PORT_ACTIVE_MTU;
~ ^~~~~~~~~~~~~~~~~~~
drivers/infiniband/sw/rxe/rxe.c:151:24: warning: implicit conversion
from enumeration type 'enum rxe_port_param' to different enumeration
type 'enum ib_mtu' [-Wenum-conversion]
ib_mtu_enum_to_int(RXE_PORT_ACTIVE_MTU);
~~~~~~~~~~~~~~~~~~ ^~~~~~~~~~~~~~~~~~~
5 warnings generated.
Use the appropriate values from the expected enumerated type so no
conversion needs to happen then remove the unneeded definitions.
Reported-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Reviewed-by: Nick Desaulniers <ndesaulniers@google.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
This patch moves ruc_loopback() from hfi1 into rdmavt for code sharing
with the qib driver.
Reviewed-by: Brian Welty <brian.welty@intel.com>
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Venkata Sandeep Dhanalakota <venkata.s.dhanalakota@intel.com>
Signed-off-by: Harish Chegondi <harish.chegondi@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Moving send completion code into rdmavt in order to have shared logic
between qib and hfi1 drivers.
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Reviewed-by: Brian Welty <brian.welty@intel.com>
Signed-off-by: Venkata Sandeep Dhanalakota <venkata.s.dhanalakota@intel.com>
Signed-off-by: Harish Chegondi <harish.chegondi@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
This patch moves hfi1_copy_sge() into rdmavt for sharing with qib.
This patch also moves all the wss_*() functions into rdmavt as
several wss_*() functions are called from hfi1_copy_sge()
When SGE copy mode is adaptive, cacheless copy may be done in some cases
for performance reasons. In those cases, X86 cacheless copy function
is called since the drivers that use rdmavt and may set SGE copy mode
to adaptive are X86 only. For this reason, this patch adds
"depends on X86_64" to rdmavt/Kconfig.
Reviewed-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
Reviewed-by: Michael J. Ruhl <michael.j.ruhl@intel.com>
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Brian Welty <brian.welty@intel.com>
Signed-off-by: Harish Chegondi <harish.chegondi@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The ll parameter is not used in ib_modify_qp_is_ok(), so remove it.
Signed-off-by: Kamal Heib <kamalheib1@gmail.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
This function is not in use - delete it.
Signed-off-by: Kamal Heib <kamalheib1@gmail.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
In rxe_queue_init, q and q->buf are allocated. In do_mmap_info, q->ip is
allocated. When error occurs, rxe_srq_from_init and the later error
handler do not free these allocated memories. This will make memory leak.
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The driver-provided function check_send_wqe allows the hardware driver to
check and set up the incoming send wqe before it is inserted into the swqe
ring. This patch will rename it as setup_wqe to better reflect its
usage. In addition, this function is only called when all setup is
complete in rdmavt.
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The buf is allocated by vmalloc_user in the function rxe_queue_init.
So it is better to free it by vfree.
Fixes: 8700e3e7c4 ("Soft RoCE driver")
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
These return the same thing but dev_name is a more conventional use of the
kernel API.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Kernel convention is that a driver for a subsystem will print using
dev_* on the subsystem's struct device, or with dev_* on the physical
device. Drivers should rarely use a pr_* function.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The current code has two copies of the device name, ibdev->dev and
dev_name(&ibdev->dev), and they are setup at different times, which is
very confusing.
Set them both up at the same time and make dev_name() the lead name, which
is the proper use of the driver core APIs. To make it very clear that the
name is not valid until registration pass it in to the
ib_register_device() call rather than messing with ibdev->name directly.
Also the reorganization now checks that dev_name is unique even if it does
not contain a %.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Acked-by: Adit Ranadive <aditr@vmware.com>
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Acked-by: Devesh Sharma <devesh.sharma@broadcom.com>
Reviewed-by: Shiraz Saleem <shiraz.saleem@intel.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Reviewed-by: Michael J. Ruhl <michael.j.ruhl@intel.com>
The post_send() path determines if it should post directly or, schedule
the post for later. The current logic is:
if the swqe ring is empty or (for hfi1) wqe->length <= piothreshold
post the send
else
schedule
This can allow large requests to call the send engine directly. Large
requests can potentially produce a large number of packets prior to
returning to the caller, blocking the caller from posting more requests,
and allowing better parallel processing.
Allow the driver(s) more say in this logic (pass call_send to the driver,
rather than examining a return value).
Update hfi1/qib logic to schedule the send engine if an RC or UC message
is larger than the QP MTU size.
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Michael J. Ruhl <michael.j.ruhl@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
When a read request is retried for the remaining partial
data, the response may restart from read response first
or read response only. So support those cases.
Do not advance the comp psn beyond the current wqe's last_psn
as that could skip over an entire read wqe and will cause the
req_retry() logic to set an incorrect req psn.
An example sequence is as follows:
Write PSN 40 -- this is the current WQE.
Read request PSN 41
Write PSN 42
Receive ACK PSN 42 -- this will complete the current WQE
for PSN 40, and set the comp psn to 42 which is a problem
because the read request at PSN 41 has been skipped over.
So when req_retry() tries to retransmit the read request,
it sets the req psn to 42 which is incorrect.
When retrying a read request, calculate the number of psns
completed based on the dma resid instead of the wqe first_psn.
The wqe first_psn could have moved if the read request was
retried multiple times.
Set the reth length to the dma resid to handle read retries for
the remaining partial data.
Signed-off-by: Vijay Immanuel <vijayi@attalasystems.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Error retries can occur due to timeouts, NAKs or receiving
packets beyond the current read request. Avoid back-to-back
retries due to packet processing, by only retrying the initial
attempt immediately. Subsequent retries must be due to timeouts.
Continue to process completion packets after scheduling a retry.
Signed-off-by: Vijay Immanuel <vijayi@attalasystems.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Don't reset the resp opcode for a replayed read response.
The resp opcode could be in the middle of a write or send
sequence, when the duplicate read request was received.
An example sequence is as follows:
- Receive read request for 12KB PSN 20. Transmit read response
first, middle and last with PSNs 20,21,22.
- Receive write first PSN 23.
At this point the resp psn is 24 and resp opcode is write first.
- The sender notices that PSN 20 is dropped and retransmits.
Receive read request for 12KB PSN 20. Transmit read response
first, middle and last with PSNs 20,21,22. The resp opcode is
set to -1, the resp psn remains 24.
- Receive write first PSN 23. This is processed by duplicate_request().
The resp opcode remains -1 and resp psn remains 24.
- Receive write middle PSN 24. check_op_seq() reports a missing
first error since the resp opcode is -1.
When sending an ack for a duplicate send or write request,
use the psn of the previous ack sent. Do not use the psn
of a read response for the ack.
An example sequence is as follows:
- Receive write PSN 30. Transmit ACK for PSN 30.
- Receive read request 4KB PSN 31. Transmit read response with
PSN 31. The resp psn is now 32.
- The sender notices that PSN 30 is dropped and retransmits.
Receive write PSN 30. duplicate_request() sends an ACK with
PSN 31. That is incorrect since PSN 31 was a read request.
Signed-off-by: Vijay Immanuel <vijayi@attalasystems.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Consolidate all error checks under single if() condition and use helper
unlikely() macro for them, in addition drop unneeded goto labels.
rxe_pool_get_index() already provides RB tree based efficient lookup.
Avoid doing extra checks for error cases which are rare and already
covered by rxe_pool_get_index().
Signed-off-by: Parav Pandit <parav@mellanox.com>
Reviewed-by: Daniel Jurgens <danielj@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
While performing lookup in a pool, if entry is found, take the
reference right there, instead of checking again outside the loop and
save one branch.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Reviewed-by: Daniel Jurgens <danielj@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Normal practice is to have enum defines in capital letters.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Reviewed-by: Daniel Jurgens <danielj@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Concurrent readers which read rb tree are protected using read lock.
Concurrent writers which add element to pool are protected
using write lock.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Reviewed-by: Daniel Jurgens <danielj@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
rxe_prepare() is called on an skb which has ndev already initialized by
rxe_init_packet().
Therefore avoid querying the GID attribute again and use the available
netdevice from the skb->dev.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Reviewed-by: Daniel Jurgens <danielj@mellanox.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Tested-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
In the commit 536ca245c5 ("IB/rxe: Drop QP0 silently"), if qpn is
zero, the function directly returns. So in the following function,
it is not necessary to check qpn. The qpn check in the function
check_keys is removed.
Fixes: 536ca245c5 ("IB/rxe: Drop QP0 silently")
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Select the source udp port number for a QP based on the
source QPN. This provides a better spread of traffic
across NIC RX queues for RC/UC QPs.
Signed-off-by: Vijay Immanuel <vijayi@attalasystems.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
-----BEGIN PGP SIGNATURE-----
iQFSBAABCAA8FiEEq68RxlopcLEwq+PEeb4+QwBBGIYFAltwm2geHHRvcnZhbGRz
QGxpbnV4LWZvdW5kYXRpb24ub3JnAAoJEHm+PkMAQRiGITkH/iSzkVhT2OxHoir0
mLVzTi7/Z17L0e/ELl7TvAC0iLFlWZKdlGR0g3b4/QpXLPmNK4HxiDRTQuWn8ke0
qDZyDq89HqLt+mpeFZ43PCd9oqV8CH2xxK3iCWReqv6bNnowGnRpSStlks4rDqWn
zURC/5sUh7TzEG4s997RrrpnyPeQWUlf/Mhtzg2/WvK2btoLWgu5qzjX1uFh3s7u
vaF2NXVJ3X03gPktyxZzwtO1SwLFS1jhwUXWBZ5AnoJ99ywkghQnkqS/2YpekNTm
wFk80/78sU+d91aAqO8kkhHj8VRrd+9SGnZ4mB2aZHwjZjGcics4RRtxukSfOQ+6
L47IdXo=
=sJkt
-----END PGP SIGNATURE-----
Merge tag 'v4.18' into rdma.git for-next
Resolve merge conflicts from the -rc cycle against the rdma.git tree:
Conflicts:
drivers/infiniband/core/uverbs_cmd.c
- New ifs added to ib_uverbs_ex_create_flow in -rc and for-next
- Merge removal of file->ucontext in for-next with new code in -rc
drivers/infiniband/core/uverbs_main.c
- for-next removed code from ib_uverbs_write() that was modified
in for-rc
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Every function that returns COMPST_ERROR must set wqe->status to another
value than IB_WC_SUCCESS before returning COMPST_ERROR. Fix the only code
path for which this is not yet the case.
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Cc: <stable@vger.kernel.org>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The proper return code is "-EOPNOTSUPP" when the create_srq() callback
is not supported.
Signed-off-by: Kamal Heib <kamalheib1@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Since neither ib_post_send() nor ib_post_recv() modify the data structure
their second argument points at, declare that argument const. This change
makes it necessary to declare the 'bad_wr' argument const too and also to
modify all ULPs that call ib_post_send(), ib_post_recv() or
ib_post_srq_recv(). This patch does not change any functionality but makes
it possible for the compiler to verify whether the
ib_post_(send|recv|srq_recv) really do not modify the posted work request.
To make this possible, only one cast had to be introduce that casts away
constness, namely in rpcrdma_post_recvs(). The only way I can think of to
avoid that cast is to introduce an additional loop in that function or to
change the data type of bad_wr from struct ib_recv_wr ** into int
(an index that refers to an element in the work request list). However,
both approaches would require even more extensive changes than this
patch.
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
When posting a send work request, the work request that is posted is not
modified by any of the RDMA drivers. Make this explicit by constifying
most ib_send_wr pointers in RDMA transport drivers.
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
According to "Annex A16: RDMA over Converged Ethernet (RoCE)":
A16.4.3 MANAGEMENT INTERFACES
As defined in the base specification, a special Queue Pair, QP0 is defined
solely for communication between subnet manager(s) and subnet management
agents. Since such an IB-defined subnet management architecture is outside
the scope of this annex, it follows that there is also no requirement that
a port which conforms to this annex be associated with a QP0. Thus, for
end nodes designed to conform to this annex, the concept of QP0 is
undefined and unused for any port connected to an Ethernet network.
CA16-8: A packet arriving at a RoCE port containing a BTH with the
destination QP field set to QP0 shall be silently dropped.
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Acked-by: Moni Shoua <monis@mellanox.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
This patch not only simplifies the error handling code in rxe_create_ah()
but also removes the dead code that was left behind by commit 47ec386662
("RDMA: Convert drivers to use sgid_attr instead of sgid_index").
Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: Parav Pandit <parav@mellanox.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Do not call sk_dst_set() on every packet transfer because
that calls sk_tx_queue_clear(), which clears the tx queue.
A QP must stay on the same tx queue to maintain packet order.
Signed-off-by: Vijay Immanuel <vijayi@attalasystems.com>
Acked-by: Moni Shoua <monis@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Regression and crashing bug fixes:
- mlx4/5: Fixes for issues found from various checkers
- A resource tracking and uverbs regression in the core code
- qedr: NULL pointer regression found during testing
- rxe: Various small bugs
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2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=Gjuy
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma
Pull rdma fixes from Jason Gunthorpe:
"Here are eight fairly small fixes collected over the last two weeks.
Regression and crashing bug fixes:
- mlx4/5: Fixes for issues found from various checkers
- A resource tracking and uverbs regression in the core code
- qedr: NULL pointer regression found during testing
- rxe: Various small bugs"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma:
IB/rxe: Fix missing completion for mem_reg work requests
RDMA/core: Save kernel caller name when creating CQ using ib_create_cq()
IB/uverbs: Fix ordering of ucontext check in ib_uverbs_write
IB/mlx4: Fix an error handling path in 'mlx4_ib_rereg_user_mr()'
RDMA/qedr: Fix NULL pointer dereference when running over iWARP without RDMA-CM
IB/mlx5: Fix return value check in flow_counters_set_data()
IB/mlx5: Fix memory leak in mlx5_ib_create_flow
IB/rxe: avoid double kfree skb
This patch replaces the ib_device_attr.max_sge with max_send_sge and
max_recv_sge. It allows ulps to take advantage of devices that have very
different send and recv sge depths. For example cxgb4 has a max_recv_sge
of 4, yet a max_send_sge of 16. Splitting out these attributes allows
much more efficient use of the SQ for cxgb4 with ulps that use the RDMA_RW
API. Consider a large RDMA WRITE that has 16 scattergather entries.
With max_sge of 4, the ulp would send 4 WRITE WRs, but with max_sge of
16, it can be done with 1 WRITE WR.
Acked-by: Sagi Grimberg <sagi@grimberg.me>
Acked-by: Christoph Hellwig <hch@lst.de>
Acked-by: Selvin Xavier <selvin.xavier@broadcom.com>
Acked-by: Shiraz Saleem <shiraz.saleem@intel.com>
Acked-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Before goto err2, the variable qp is checked. So it is not necessary
to check qp in label err2.
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Set the vlan flag and vlan_id field in the wc for rdma_listen()
to work over VLAN. This is required by ib_init_ah_attr_from_wc()
which is called by the CM REQ handler.
Signed-off-by: Vijay Immanuel <vijayi@attalasystems.com>
Reviewed-by: Yonatan Cohen <yonatanc@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Increase the max MR limit to support more I/O queues
for NVMe over Fabrics hosts.
Signed-off-by: Vijay Immanuel <vijayi@attalasystems.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Run the completer task to post a work completion after processing
a memory registration or invalidate work request. This covers the
case where the memory registration or invalidate was the last work
request posted to the qp.
Signed-off-by: Vijay Immanuel <vijayi@attalasystems.com>
Reviewed-by: Yonatan Cohen <yonatanc@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The core code now ensures that all driver callbacks that receive an
rdma_ah_attrs will have a sgid_attr's pointer if there is a GRH present.
Drivers can use this pointer instead of calling a query function with
sgid_index. This simplifies the drivers and also avoids races where a
gid_index lookup may return different data if it is changed.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Introduce AH attribute copy, move and replace APIs to be used by core and
provider drivers.
In CM code flow when ah attribute might be re-initialized twice while
processing incoming request, or initialized once while from path record
while sending out CM requests. Therefore use rdma_move_ah_attr API to
handle such scenarios instead of memcpy().
Provider drivers keeps a copy ah_attr during the lifetime of the ah.
Therefore, use rdma_replace_ah_attr() which conditionally release
reference to old ah_attr and holds reference to new attribute whose
referrence is released when the AH is freed.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
struct rxe_global_route and struct ib_global_route are not the same thing
and should not be memcpy'd over each other, do a member by member copy
instead. This allows the layout of the in-kernel struct ib_global_route to
be changed without breaking rxe.
Reviewed-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
rxe_netdev_from_av can now be done by the core code directly from the
gid_attrs, no need for a helper in the driver.
ib_find_cached_gid_by_port can be switched to use the rdma version here as
well.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
In rxe_send, when network_type is not RDMA_NETWORK_IPV4 or
RDMA_NETWORK_IPV6, skb is freed and -EINVAL is returned.
Then rxe_xmit_packet will return -EINVAL, too. In rxe_requester,
this skb is double freed.
In rxe_requester, kfree_skb is needed only after fill_packet fails.
So kfree_skb is moved from label err to test fill_packet.
Fixes: 5793b46521 ("IB/rxe: remove unnecessary skb_clone in xmit")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
This has been a quiet cycle for RDMA, the big bulk is the usual smallish
driver updates and bug fixes. About four new uAPI related things. Not as much
Szykaller patches this time, the bugs it finds are getting harder to fix.
- More work cleaning up the RDMA CM code
- Usual driver bug fixes and cleanups for qedr, qib, hfi1, hns, i40iw, iw_cxgb4, mlx5, rxe
- Driver specific resource tracking and reporting via netlink
- Continued work for name space support from Parav
- MPLS support for the verbs flow steering uAPI
- A few tricky IPoIB fixes improving robustness
- HFI1 driver support for the '16B' management packet format
- Some auditing to not print kernel pointers via %llx or similar
- Mark the entire 'UCM' user-space interface as BROKEN with the intent to remove it
entirely. The user space side of this was long ago replaced with RDMA-CM and
syzkaller is finding bugs in the residual UCM interface nobody wishes to fix because
nobody uses it.
- Purge more bogus BUG_ON's from Leon
- 'flow counters' verbs uAPI
- T10 fixups for iser/isert, these are Acked by Martin but going through the RDMA
tree due to dependencies
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2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=Mb8m
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma
Pull rdma updates from Jason Gunthorpe:
"This has been a quiet cycle for RDMA, the big bulk is the usual
smallish driver updates and bug fixes. About four new uAPI related
things. Not as much Szykaller patches this time, the bugs it finds are
getting harder to fix.
Summary:
- More work cleaning up the RDMA CM code
- Usual driver bug fixes and cleanups for qedr, qib, hfi1, hns,
i40iw, iw_cxgb4, mlx5, rxe
- Driver specific resource tracking and reporting via netlink
- Continued work for name space support from Parav
- MPLS support for the verbs flow steering uAPI
- A few tricky IPoIB fixes improving robustness
- HFI1 driver support for the '16B' management packet format
- Some auditing to not print kernel pointers via %llx or similar
- Mark the entire 'UCM' user-space interface as BROKEN with the
intent to remove it entirely. The user space side of this was long
ago replaced with RDMA-CM and syzkaller is finding bugs in the
residual UCM interface nobody wishes to fix because nobody uses it.
- Purge more bogus BUG_ON's from Leon
- 'flow counters' verbs uAPI
- T10 fixups for iser/isert, these are Acked by Martin but going
through the RDMA tree due to dependencies"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (138 commits)
RDMA/mlx5: Update SPDX tags to show proper license
RDMA/restrack: Change SPDX tag to properly reflect license
IB/hfi1: Fix comment on default hdr entry size
IB/hfi1: Rename exp_lock to exp_mutex
IB/hfi1: Add bypass register defines and replace blind constants
IB/hfi1: Remove unused variable
IB/hfi1: Ensure VL index is within bounds
IB/hfi1: Fix user context tail allocation for DMA_RTAIL
IB/hns: Use zeroing memory allocator instead of allocator/memset
infiniband: fix a possible use-after-free bug
iw_cxgb4: add INFINIBAND_ADDR_TRANS dependency
IB/isert: use T10-PI check mask definitions from core layer
IB/iser: use T10-PI check mask definitions from core layer
RDMA/core: introduce check masks for T10-PI offload
IB/isert: fix T10-pi check mask setting
IB/mlx5: Add counters read support
IB/mlx5: Add flow counters read support
IB/mlx5: Add flow counters binding support
IB/mlx5: Add counters create and destroy support
IB/uverbs: Add support for flow counters
...
- Use overflow helpers in 2-factor allocators (Kees, Rasmus)
- Introduce overflow test module (Rasmus, Kees)
- Introduce saturating size helper functions (Matthew, Kees)
- Treewide use of struct_size() for allocators (Kees)
-----BEGIN PGP SIGNATURE-----
Comment: Kees Cook <kees@outflux.net>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=/3L8
-----END PGP SIGNATURE-----
Merge tag 'overflow-v4.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux
Pull overflow updates from Kees Cook:
"This adds the new overflow checking helpers and adds them to the
2-factor argument allocators. And this adds the saturating size
helpers and does a treewide replacement for the struct_size() usage.
Additionally this adds the overflow testing modules to make sure
everything works.
I'm still working on the treewide replacements for allocators with
"simple" multiplied arguments:
*alloc(a * b, ...) -> *alloc_array(a, b, ...)
and
*zalloc(a * b, ...) -> *calloc(a, b, ...)
as well as the more complex cases, but that's separable from this
portion of the series. I expect to have the rest sent before -rc1
closes; there are a lot of messy cases to clean up.
Summary:
- Introduce arithmetic overflow test helper functions (Rasmus)
- Use overflow helpers in 2-factor allocators (Kees, Rasmus)
- Introduce overflow test module (Rasmus, Kees)
- Introduce saturating size helper functions (Matthew, Kees)
- Treewide use of struct_size() for allocators (Kees)"
* tag 'overflow-v4.18-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux:
treewide: Use struct_size() for devm_kmalloc() and friends
treewide: Use struct_size() for vmalloc()-family
treewide: Use struct_size() for kmalloc()-family
device: Use overflow helpers for devm_kmalloc()
mm: Use overflow helpers in kvmalloc()
mm: Use overflow helpers in kmalloc_array*()
test_overflow: Add memory allocation overflow tests
overflow.h: Add allocation size calculation helpers
test_overflow: Report test failures
test_overflow: macrofy some more, do more tests for free
lib: add runtime test of check_*_overflow functions
compiler.h: enable builtin overflow checkers and add fallback code
One of the more common cases of allocation size calculations is finding
the size of a structure that has a zero-sized array at the end, along
with memory for some number of elements for that array. For example:
struct foo {
int stuff;
void *entry[];
};
instance = kmalloc(sizeof(struct foo) + sizeof(void *) * count, GFP_KERNEL);
Instead of leaving these open-coded and prone to type mistakes, we can
now use the new struct_size() helper:
instance = kmalloc(struct_size(instance, entry, count), GFP_KERNEL);
This patch makes the changes for kmalloc()-family (and kvmalloc()-family)
uses. It was done via automatic conversion with manual review for the
"CHECKME" non-standard cases noted below, using the following Coccinelle
script:
// pkey_cache = kmalloc(sizeof *pkey_cache + tprops->pkey_tbl_len *
// sizeof *pkey_cache->table, GFP_KERNEL);
@@
identifier alloc =~ "kmalloc|kzalloc|kvmalloc|kvzalloc";
expression GFP;
identifier VAR, ELEMENT;
expression COUNT;
@@
- alloc(sizeof(*VAR) + COUNT * sizeof(*VAR->ELEMENT), GFP)
+ alloc(struct_size(VAR, ELEMENT, COUNT), GFP)
// mr = kzalloc(sizeof(*mr) + m * sizeof(mr->map[0]), GFP_KERNEL);
@@
identifier alloc =~ "kmalloc|kzalloc|kvmalloc|kvzalloc";
expression GFP;
identifier VAR, ELEMENT;
expression COUNT;
@@
- alloc(sizeof(*VAR) + COUNT * sizeof(VAR->ELEMENT[0]), GFP)
+ alloc(struct_size(VAR, ELEMENT, COUNT), GFP)
// Same pattern, but can't trivially locate the trailing element name,
// or variable name.
@@
identifier alloc =~ "kmalloc|kzalloc|kvmalloc|kvzalloc";
expression GFP;
expression SOMETHING, COUNT, ELEMENT;
@@
- alloc(sizeof(SOMETHING) + COUNT * sizeof(ELEMENT), GFP)
+ alloc(CHECKME_struct_size(&SOMETHING, ELEMENT, COUNT), GFP)
Signed-off-by: Kees Cook <keescook@chromium.org>
Update mlx4 to support user MR creation against read-only memory, previously
it required the memory to be writable.
Based on rdma for-rc due to dependencies.
* mr_fix: (2 commits)
IB/mlx4: Mark user MR as writable if actual virtual memory is writable
IB/core: Make testing MR flags for writability a static inline function
The function rxe_remove_all is only used in this modules.
There is no other modules that call this function. So it
is not necessary to export it.
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Given we are dealing with nano-second level timers, when the timer
pops, ensure it happens on the CPU which caused the timer to be set
in the first place. This avoids excessive jitter from the desired
expiration time by avoiding the cost of switching our context to
another CPU that is cache cold for this given timer.
Reviewed-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
In the exit branch, WARN_ON_ONCE is called to show stack. So it is
not necessary to call WARN_ON_ONCE before going to exit.
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
DMA_VIRT_OPS requires that dma_addr_t is at least as wide as a
pointer, which is expressed as a dependency on !64BIT ||
ARCH_DMA_ADDR_T_64BIT.
For parisc64 this is not true, and if these IB modules are enabled,
kconfig warns:
WARNING: unmet direct dependencies detected for DMA_VIRT_OPS
Depends on [n]: HAS_DMA [=y] && (!64BIT [=y] || ARCH_DMA_ADDR_T_64BIT)
Selected by [m]:
- INFINIBAND_RDMAVT [=m] && INFINIBAND [=m] && 64BIT [=y] && PCI [=y]
- RDMA_RXE [=m] && INET [=y] && PCI [=y] && INFINIBAND [=m]
Add dependencies to fix this.
Signed-off-by: Ben Hutchings <ben@decadent.org.uk>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Moving receive-side WQE allocation logic into rdmavt will allow
further code reuse between qib and hfi1 drivers.
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Brian Welty <brian.welty@intel.com>
Signed-off-by: Harish Chegondi <harish.chegondi@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Currently the driver doesn't support completion vectors. These
are used to indicate which sets of CQs should be grouped together
into the same vector. A vector is a CQ processing thread that
runs on a specific CPU.
If an application has several CQs bound to different completion
vectors, and each completion vector runs on different CPUs, then
the completion queue workload is balanced. This helps scale as more
nodes are used.
Implement CQ completion vector support using a global workqueue
where a CQ entry is queued to the CPU corresponding to the CQ's
completion vector. Since the workqueue is global, it's guaranteed
to always be there when queueing CQ entries; Therefore, the RCU
locking for cq->rdi->worker in the hot path is superfluous.
Each completion vector is assigned to a different CPU. The number of
completion vectors available is computed by taking the number of
online, physical CPUs from the local NUMA node and subtracting the
CPUs used for kernel receive queues and the general interrupt.
Special use cases:
* If there are no CPUs left for completion vectors, the same CPU
for the general interrupt is used; Therefore, there would only
be one completion vector available.
* For multi-HFI systems, the number of completion vectors available
for each device is the total number of completion vectors in
the local NUMA node divided by the number of devices in the same
NUMA node. If there's a division remainder, the first device to
get initialized gets an extra completion vector.
Upon a CQ creation, an invalid completion vector could be specified.
Handle it as follows:
* If the completion vector is less than 0, set it to 0.
* Set the completion vector to the result of the passed completion
vector moded with the number of device completion vectors
available.
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Sebastian Sanchez <sebastian.sanchez@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
All threads queuing CQ entries on different CQs are unnecessarily
synchronized by a spin lock to check if the CQ kthread worker hasn't
been destroyed before queuing an CQ entry.
The lock used in 6efaf10f16 ("IB/rdmavt: Avoid queuing work into a
destroyed cq kthread worker") is a device global lock and will have
poor performance at scale as completions are entered from a large
number of CPUs.
Convert to use RCU where the read side of RCU is rvt_cq_enter() to
determine that the worker is alive prior to triggering the
completion event.
Apply write side RCU semantics in rvt_driver_cq_init() and
rvt_cq_exit().
Fixes: 6efaf10f16 ("IB/rdmavt: Avoid queuing work into a destroyed cq kthread worker")
Cc: <stable@vger.kernel.org> # 4.14.x
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Sebastian Sanchez <sebastian.sanchez@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Several items of conflict have arisen between the RDMA stack's for-rc
branch and upcoming for-next work:
9fd4350ba8 ("IB/rxe: avoid double kfree_skb") directly conflicts with
2e47350789 ("IB/rxe: optimize the function duplicate_request")
Patches already submitted by Intel for the hfi1 driver will fail to
apply cleanly without this merge
Other people on the mailing list have notified that their upcoming
patches also fail to apply cleanly without this merge
Signed-off-by: Doug Ledford <dledford@redhat.com>
Use the recently introduced helper to replace the pattern of
skb_put_zero/__skb_put() && memset().
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Reviewed-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
When skb is sent, it will pass the following functions in soft roce.
rxe_send [rdma_rxe]
ip_local_out
__ip_local_out
ip_output
ip_finish_output
ip_finish_output2
dev_queue_xmit
__dev_queue_xmit
dev_hard_start_xmit
In the above functions, if error occurs in the above functions or
iptables rules drop skb after ip_local_out, kfree_skb will be called.
So it is not necessary to call kfree_skb in soft roce module again.
Or else crash will occur.
The steps to reproduce:
server client
--------- ---------
|1.1.1.1|<----rxe-channel--->|1.1.1.2|
--------- ---------
On server: rping -s -a 1.1.1.1 -v -C 10000 -S 512
On client: rping -c -a 1.1.1.1 -v -C 10000 -S 512
The kernel configs CONFIG_DEBUG_KMEMLEAK and
CONFIG_DEBUG_OBJECTS are enabled on both server and client.
When rping runs, run the following command in server:
iptables -I OUTPUT -p udp --dport 4791 -j DROP
Without this patch, crash will occur.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
w/o RXE_START_MASK, the last_psn of IB_OPCODE_RC_SEND_ONLY_INV
will not be updated in update_wqe_psn, and the corresponding
wqe will not be acked in rxe_completer due to its last_psn is
zero. Finally, the other wqe will also not be able to be acked,
because the wqe of IB_OPCODE_RC_SEND_ONLY_INV with last_psn 0
is still there. This causes large amount of io timeout when
nvmeof is over rxe.
Add RXE_START_MASK for IB_OPCODE_RC_SEND_ONLY_INV to fix this.
Signed-off-by: Jianchao Wang <jianchao.w.wang@oracle.com>
Reviewed-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
In the functions rxe_mem_init_dma, rxe_mem_init_user, rxe_mem_init_fast
and copy_data, the function variable rxe is not used. So this function
variable rxe is removed.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
The function rxe_set_mtu always returns zero. So this function type
is changed to void.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Follow the advice from Bart, the function refcount_inc is replaced
with skb_get in commit 99dae69025 ("IB/rxe: optimize mcast recv process")
and commit 86af617641 ("IB/rxe: remove unnecessary skb_clone").
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Suggested-by: Bart Van Assche <Bart.VanAssche@wdc.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
In the function duplicate_request, the reference of skb can be increased
to replace the function skb_clone.
This will make rxe performace better and save memory.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Bart Van Assche <bart.vanassche@wdc.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
The function rxe_release_udp_tunnel is only used in rxe_net.c.
So it is necessary to make this function as static.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
The functions rxe_set_mtu, rxe_add and rxe_remove are only used in their
own module. So it is not necessary to export them.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
The variable rxe_net_notifier is only used in the file rxe_net.c. So
remove it from rxe_net.h file and make it static in the file rxe_net.c.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Bart Van Assche <bart.vanassche@wdc.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
rxe driver's add_gid() and del_gid() callbacks are doing simple
checks which are already done by the ib core before invoking these
callback routines.
Therefore, code is simplified to skip implementing add_gid() and
del_gid() callback functions.
They are only invoked by ib_core if they are implemented.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Now that ib_gid_attr contains device, port and index, simplify the
provider APIs add_gid() and del_gid() to use device, port and index
fields from the ib_gid_attr attributes structure.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Now that IB core GID cache ensures that all RoCE entries have an
associated netdev remove null checks from the provider drivers for
clarity.
Reviewed-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
ib_query_gid() fetches the GID from the software cache maintained in
ib_core for RoCE ports.
Therefore, simplify the provider drivers for RoCE to treat query_gid()
callback as never called for RoCE, and only require non-RoCE devices to
implement it.
Signed-off-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
In mcast recv process, the function skb_clone is used. In fact,
the refcount can be increased to replace cloning a new skb since
the original skb will not be modified before it is freed.
This can make the performance better and save the memory.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
With 32 bit compilation several of the fields become misaligned here.
Fixing this is an ABI break for 32 bit rxe and it is in well used
portions of the rxe ABI.
To handle this we bump the ABI version, as expected. However the user
space driver doesn't handle it properly today, so all existing user
space continues to work.
Updated userspace will start to require the necessary kernel version.
We don't expect there to be any 32 bit users of rxe. Most likely cases,
such as ARM 32 already generally don't work because rxe does not handle
the CPU cache properly on its shared with userspace pages.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Extending uverbs_ioctl header with driver_id and another reserved
field. driver_id should be used in order to identify the driver.
Since every driver could have its own parsing tree, this is necessary
for strace support.
Downstream patches take off the EXPERIMENTAL flag from the ioctl() IB
support and thus we add some reserved fields for future usage.
Reviewed-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Matan Barak <matanb@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Open coding pointer math is not acceptable for describing the uABI in
RDMA. Provide structs for all the cases.
The udata is casted to the struct as close to the verbs entry point
as possible for maximum clarity. Function signatures and so forth
are revised to allow for this.
Tested-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
It isn't used and it couldn't possibly ever be used correctly.
Tested-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
rvt_mregion uses percpu_ref for reference counting and RCU to protect
accesses from lkey_table. When a rvt_mregion needs to be freed, it
first gets unregistered from lkey_table and then rvt_check_refs() is
called to wait for in-flight usages before the rvt_mregion is freed.
rvt_check_refs() seems to have a couple issues.
* It has a fast exit path which tests percpu_ref_is_zero(). However,
a percpu_ref reading zero doesn't mean that the object can be
released. In fact, the ->release() callback might not even have
started executing yet. Proceeding with freeing can lead to
use-after-free.
* lkey_table is RCU protected but there is no RCU grace period in the
free path. percpu_ref uses RCU internally but it's sched-RCU whose
grace periods are different from regular RCU. Also, it generally
isn't a good idea to depend on internal behaviors like this.
To address the above issues, this patch removes the fast exit and adds
an explicit synchronize_rcu().
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Cc: Mike Marciniszyn <mike.marciniszyn@intel.com>
Cc: linux-rdma@vger.kernel.org
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
This patch fixes RDMA/rxe over 802.1q VLAN devices.
Without it, I observed the following behavior:
a) adding a VLAN device to RXE via rxe_net_add() creates a non-functional
RDMA device. This is caused by the logic in enum_all_gids_of_dev_cb() /
is_eth_port_of_netdev(), which only considers networks connected to
"upper devices" of the configured network device, resulting in an empty
set of gids for a VLAN interface that is an "upper device" itself.
Later attempts to connect via this rdma device fail in cma_acuire_dev()
because no gids can be resolved.
b) adding the master device of the VLAN device instead seems to work
initially, target addresses via VLAN devices are resolved successfully.
But the connection times out because no 802.1q VLAN headers are
inserted in the ethernet packets, which are therefore never received.
This happens because the RXE layer sends the packets via the master
device rather than the VLAN device.
The problem could be solved by changing either a) or b). My thinking was
that the logic in a) was created deliberately, thus I decided to work on
b). It turns out that the information about the VLAN interface for the gid
at hand is available in the AV information. My patch converts the RXE code
to use this netdev instead of rxe->ndev. With this change, RXE over vlan
works on my test system.
Signed-off-by: Martin Wilck <mwilck@suse.com>
Reviewed-by: Moni Shoua <monis@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
In hfi.h, the header file opa_addr.h is included twice.
In vt.h, the header file mmap.h is included twice.
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Acked-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
The function rxe_init_device_param always return 0. So the function
type is changed to void.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
In the function rxe_send, the variable rxe is not used in it.
So it should be removed.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
In send_atomic_ack function, it is not necessary to make a
skb_clone. To gain better performance (high throughput and
low latency), this skb_clone is removed.
The following tests are made.
server client
--------- ---------
|1.1.1.1|<----rxe-channel--->|1.1.1.2|
--------- ---------
On server: rping -s -a 1.1.1.1 -v -C 1000 -S 512
On client: rping -c -a 1.1.1.1 -v -C 1000 -S 512
The kernel config CONFIG_DEBUG_KMEMLEAK is enabled on both server
and client.
This test runs for several hours. There is no memory leak and the whole
system can work well.
Based on the above network, the following tests are made.
Server: ibv_rc_pingpong -d rxe0 -g 1
Client: ibv_rc_pingpong -d rxe0 -g 1 1.1.1.1
The test results on Server(10 tests are made).
Before:
Throughput is 137.07 Mbit/sec
Latency is 517.76 usec/iter
After:
Throughput is 148.85 Mbit/sec
Latency is 476.64 usec/iter
The throughput is enhanced and the latency is reduced.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
The pointer rdi is being initialized with a value that is never read
and re-assigned immediately after, hence the initialization is redundant
and can be removed.
Cleans up clang warning:
drivers/infiniband/sw/rdmavt/vt.c:94:23: warning: Value stored to 'rdi'
during its initialization is never read
Signed-off-by: Colin Ian King <colin.king@canonical.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Note: This is compile only tested as I have no access to the hw. This
variable was not used anywhere in the code. Removing it saves 24 bytes.
add/remove: 0/1 grow/shrink: 0/0 up/down: 0/-24 (-24)
Function old new delta
rxe_qp_state_name 24 - -24
Total: Before=3348732, After=3348708, chg -0.00%
Signed-off-by: Hernán Gonzalez <hernan@vanguardiasur.com.ar>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Linux has two 'linux/socket.h' files - and only the one in the kernel
defines struct sockaddr - the user space one does not.
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
- Clean up some function signatures in rxe for clarity
- Tidy the RDMA netlink header to remove unimplemented constants
- bnxt_re driver fixes, one is a regression this window.
- Minor hns driver fixes
- Various fixes from Dan Carpenter and his tool
- Fix IRQ cleanup race in HFI1
- HF1 performance optimizations and a fix to report counters in the right units
- Fix for an IPoIB startup sequence race with the external manager
- Oops fix for the new kabi path
- Endian cleanups for hns
- Fix for mlx5 related to the new automatic affinity support
-----BEGIN PGP SIGNATURE-----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=xwZz
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma
Pull more rdma updates from Doug Ledford:
"Items of note:
- two patches fix a regression in the 4.15 kernel. The 4.14 kernel
worked fine with NVMe over Fabrics and mlx5 adapters. That broke in
4.15. The fix is here.
- one of the patches (the endian notation patch from Lijun) looks
like a lot of lines of change, but it's mostly mechanical in
nature. It amounts to the biggest chunk of change in it (it's about
2/3rds of the overall pull request).
Summary:
- Clean up some function signatures in rxe for clarity
- Tidy the RDMA netlink header to remove unimplemented constants
- bnxt_re driver fixes, one is a regression this window.
- Minor hns driver fixes
- Various fixes from Dan Carpenter and his tool
- Fix IRQ cleanup race in HFI1
- HF1 performance optimizations and a fix to report counters in the right units
- Fix for an IPoIB startup sequence race with the external manager
- Oops fix for the new kabi path
- Endian cleanups for hns
- Fix for mlx5 related to the new automatic affinity support"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (38 commits)
net/mlx5: increase async EQ to avoid EQ overrun
mlx5: fix mlx5_get_vector_affinity to start from completion vector 0
RDMA/hns: Fix the endian problem for hns
IB/uverbs: Use the standard kConfig format for experimental
IB: Update references to libibverbs
IB/hfi1: Add 16B rcvhdr trace support
IB/hfi1: Convert kzalloc_node and kcalloc to use kcalloc_node
IB/core: Avoid a potential OOPs for an unused optional parameter
IB/core: Map iWarp AH type to undefined in rdma_ah_find_type
IB/ipoib: Fix for potential no-carrier state
IB/hfi1: Show fault stats in both TX and RX directions
IB/hfi1: Remove blind constants from 16B update
IB/hfi1: Convert PortXmitWait/PortVLXmitWait counters to flit times
IB/hfi1: Do not override given pcie_pset value
IB/hfi1: Optimize process_receive_ib()
IB/hfi1: Remove unnecessary fecn and becn fields
IB/hfi1: Look up ibport using a pointer in receive path
IB/hfi1: Optimize packet type comparison using 9B and bypass code paths
IB/hfi1: Compute BTH only for RDMA_WRITE_LAST/SEND_LAST packet
IB/hfi1: Remove dependence on qp->s_hdrwords
...
In the function rxe_av_fill_ip_info, the parameter rxe is not used.
So it is removed.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
The function rxe_av_fill_ip_info always returns 0. So the function
type is changed to void.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Since the function rxe_av_to_attr always return 0, the function
type is changed to void.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
In the function rxe_av_to_attr, the parameter rxe is
not used. So it is removed.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
The function rxe_av_from_attr always return 0. So change the
function to void.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
In the function rxe_av_from_attr, the parameter rxe
is not used. So it is removed.
CC: Srinivas Eeda <srinivas.eeda@oracle.com>
CC: Junxiao Bi <junxiao.bi@oracle.com>
Signed-off-by: Zhu Yanjun <yanjun.zhu@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Yuval Shaia <yuval.shaia@oracle.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Andrew Boyer