Active QSFP cables were reset only every alternate iteration of the
channel tuning algorithm instead of every iteration due to incorrect
reset of the flag that controlled QSFP reset, resulting in using stale
QSFP status in the channel tuning algorithm.
Fixes: 8ebd4cf185 ("Add active and optical cable support")
Reviewed-by: Dean Luick <dean.luick@intel.com>
Signed-off-by: Easwar Hariharan <easwar.hariharan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Some QSFP cables assert the interrupt line as a side effect of module
plug-in and power up. This causes the SerDes and QSFP tuning algorithm
to begin cable initialization by reading the QSFP memory map over I2C,
which fails. This patch ignores any interrupt line assertion until
the module has completed power up and voltage rails have stabilized,
which can take a maximum of 500 ms per the SFF-8679 specification.
Reviewed-by: Dean Luick <dean.luick@intel.com>
Signed-off-by: Easwar Hariharan <easwar.hariharan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
QSFP CDR enablement is now controlled by determining power class
and the configuration file. We disable the DC 8051 from requesting
enablement or disabling of TX and RX CDRs by removing the code
that allowed the DC 8051 to request changes.
Reviewed-by: Dean Luick <dean.luick@intel.com>
Signed-off-by: Easwar Hariharan <easwar.hariharan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Hanging has been observed while writing a file over NFSoRDMA. Dmesg on
the server contains messages like these:
[ 931.992501] svcrdma: Error -22 posting RDMA_READ
[ 952.076879] svcrdma: Error -22 posting RDMA_READ
[ 982.154127] svcrdma: Error -22 posting RDMA_READ
[ 1012.235884] svcrdma: Error -22 posting RDMA_READ
[ 1042.319194] svcrdma: Error -22 posting RDMA_READ
Here is why:
With the base memory management extension enabled, FRMR is used instead
of FMR. The xprtrdma server issues each RDMA read request as the following
bundle:
(1)IB_WR_REG_MR, signaled;
(2)IB_WR_RDMA_READ, signaled;
(3)IB_WR_LOCAL_INV, signaled & fencing.
These requests are signaled. In order to generate completion, the fast
register work request is processed by the hfi1 send engine after being
posted to the work queue, and the corresponding lkey is not valid until
the request is processed. However, the rdmavt driver validates lkey when
the RDMA read request is posted and thus it fails immediately with error
-EINVAL (-22).
This patch changes the work flow of local operations (fast register and
local invalidate) so that fast register work requests are always
processed immediately to ensure that the corresponding lkey is valid
when subsequent work requests are posted. Local invalidate requests are
processed immediately if fencing is not required and no previous local
invalidate request is pending.
To allow completion generation for signaled local operations that have
been processed before posting to the work queue, an internal send flag
RVT_SEND_COMPLETION_ONLY is added. The hfi1 send engine checks this flag
and only generates completion for such requests.
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Jianxin Xiong <jianxin.xiong@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Enhance the PCIe Gen3 recipe to support static CTLE tuning,
and add a switch to choose between static and dynamic
approaches. Make discrete chips default to static CTLE
tuning.
Reviewed-by: Tadeusz Struk <tadeusz.struk@intel.com>
Signed-off-by: Dean Luick <dean.luick@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
This fixes the following warnings with PROVE_LOCKING and PROVE_RCU
enabled in the kernel:
case (1):
[ INFO: suspicious RCU usage. ]
drivers/infiniband/hw/hfi1/init.c:532
suspicious rcu_dereference_check() usage!
case (2):
[ INFO: suspicious RCU usage. ]
drivers/infiniband/hw/hfi1/hfi.h:1624
suspicious rcu_dereference_check() usage!
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Jianxin Xiong <jianxin.xiong@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
When link up fails in LNI, the local and peer state complete
frames are reported as numbers. Explain what the values mean
so the operator can better diagnose the problem.
Reviewed-by: Easwar Hariharan <easwar.hariharan@intel.com>
Signed-off-by: Dean Luick <dean.luick@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Currently, the default number of kernel receive contexts is set to the
number of NUMA nodes on the system plus one for control context. However,
the systems that have a single socket and/or have NUMA disabled in the BIOS
will have only one receive context by default. This patch would ensure that
by default there will be at least two kernel receive contexts plus one for
control context regardless of the number of NUMA nodes on the system. The
user can override the default number of kernel receive contexts with the
krcvqs module parameter.
Reviewed-by: Dean Luick <dean.luick@intel.com>
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@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>
In order to support extended memory management support, add send side
processing of work requests of type IB_WR_REG_MR, IB_WR_LOCAL_INV, and
IB_WR_SEND_WITH_INV. The first two are local operations and are supported
for both RC and UC. Send with invalidate is only supported for RC because
the corresponding IB opcodes are not defined for UC.
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jianxin Xiong <jianxin.xiong@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
There were multiple places where FECN/BECN processing was
being done for the different types of QPs. All of that code
was very similar, which meant that it could be pulled into
a single function used by the different QP types.
To retain the performance in the fastpath, the common code
starts with an inline function, which only calls the slow
path if the packet has any of the [FB]ECN bits set.
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Mitko Haralanov <mitko.haralanov@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Currently each user context is assigned a single SDMA engine
based on the VL, context id, and subcontext id. That means for
MPI applications, each rank can only use one SDMA engine for
all messages. This may create unwanted backup for independent
messages going to different destinations upon congestion at one
destination.
This patch adds the packet "dlid" to the formula of SDMA engine
selection for user SDMA requests. A simple hash table is used
to maintain even distribution among the available SDMA engines
regardless how the "dlid" values are distributed.
Reviewed-by: Dean Luick <dean.luick@intel.com>
Reviewed-by: Tadeusz Struk <tadeusz.struk@intel.com>
Signed-off-by: Jianxin Xiong <jianxin.xiong@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
When performing process affinity recommendations for MPI ranks, the current
algorithm doesn't take into account multiple HFI units. Also, real
cores and HT cores are not distinguished from one another. Therefore,
all HT cores are recommended to be assigned first within the local NUMA
node before recommending the assignments of cores in other NUMA nodes.
It's ideal to assign all real cores across all NUMA nodes first, then all
HT 1 cores, then all HT 2 cores, and so on to balance CPU workload. CPU
cores in other NUMA nodes could be running interrupt handlers, and this is
not taken into account.
To balance the CPU workload for user processes, the following
recommendation algorithm is used:
For each user process that is opening a context on HFI Y:
a) If all cores are assigned to user processes, start assignments all
over from the first core
b) Assign real cores first, then HT cores (First set of HT cores on
all physical cores, then second set of HT cores, and, so on) in the
following order:
1. Same NUMA node as HFI Y and not running an IRQ handler
2. Same NUMA node as HFI Y and running an IRQ handler
3. Different NUMA node to HFI Y and not running an IRQ handler
4. Different NUMA node to HFI Y and running an IRQ handler
c) Mark core as assigned in the global affinity structure. As user
processes are done, remove core assignments from global affinity
structure.
This implementation allows an arbitrary number of HT cores and provides
support for multiple HFIs.
This is being included in the kernel rather than user space due to the
fact that user space has no way of knowing the CPU recommendations for
contexts running as part of other jobs.
Reviewed-by: Ira Weiny <ira.weiny@intel.com>
Reviewed-by: Mitko Haralanov <mitko.haralanov@intel.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Sebastian Sanchez <sebastian.sanchez@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Kernel receive queues oversubscribe CPU cores on multi-HFI systems.
To prevent this, the kernel receive queues are separated onto
different cores, and the SDMA engine interrupts are constrained to
a lesser number of cores.
hfi1s_on_numa_node*krcvqs is the number of CPU cores that are
reserved for kernel receive queues for all HFIs. Each HFI initializes
its kernel receive queues to one of the reserved CPU cores. If there
ends up being 0 CPU cores leftover for SDMA engines, use the same
CPU cores as receive contexts.
In addition, general and control contexts are assigned to their own
CPU core, however, both types of contexts tend to have low traffic.
To save CPU cores, collapse general and control contexts to one CPU
core for all HFI units. This change prevents SDMA engine interrupts
from wrapping around general contexts.
Reviewed-by: Dean Luick <dean.luick@intel.com>
Signed-off-by: Sebastian Sanchez <sebastian.sanchez@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
When HFI units get initialized, they each use their own mask copy for
affinity assignments. On a multi-HFI system, affinity assignments
overbook CPU cores as each HFI doesn't have knowledge of affinity
assignments for other HFI units. Therefore, some CPU cores are never
used for interrupt handlers in systems with high number of CPU cores
per NUMA node.
For multi-HFI systems, SDMA engine interrupt assignments start all over
from the first CPU in the local NUMA node after the first HFI
initialization. This change allows assignments to continue where the
last HFI unit left off.
Add global structure for affinity assignments for multiple HFIs to share
affinity mask.
Reviewed-by: Jianxin Xiong <jianxin.xiong@intel.com>
Reviewed-by: Jubin John <jubin.john@intel.com>
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Sebastian Sanchez <sebastian.sanchez@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
The q-counter-id is given in modify-QP command associates
the QP with the counter. The offset to which the counter
ID was set is incorrect, causing IB port counters not to
count on QP.
Fixes: 0837e86a7a ('IB/mlx5: Add per port counters')
Signed-off-by: Alex Vesker <valex@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Tested-by: Mark Bloch <markb@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Number of outstanding_pi may overflow and as a result may indicate that
there are no elements in the queue. The effect of doing this is that the
MAD layer will get stuck waiting for completions. The MAD layer will
think that the QP is full - because it didn't receive these completions.
This fix changes it so the outstanding_pi number is increased
with 32-bit wraparound and is not limited to max_send_wr so
that the difference between outstanding_pi and outstanding_ci will
really indicate the number of outstanding completions.
Cc: Stable <stable@vger.kernel.org>
Fixes: ea6dc20362 ('IB/mlx5: Reorder GSI completions')
Signed-off-by: Slava Shwartsman <slavash@mellanox.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Reviewed-by: Haggai Eran <haggaie@mellanox.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Doug Ledford <dledford@redhat.com>
i40iw_puda_get_listbuf may return NULL if the list is empty.
Add NULL check prior to accessing the pointer.
Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com>
Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
In i40iw_cq_poll_completion, we always move the tail. So there is
no reason to check for overflow everytime we move the head.
Signed-off-by: Mustafa Ismail <mustafa.ismail@intel.com>
Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Blocking in c4iw_destroy_qp() causes a deadlock when apps destroy a qp
or disconnect a cm_id from their cm event handler function. There is
no need to block here anyway, so just replace the refcnt atomic with a
kref object and free the memory on the last put.
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Doug Ledford <dledford@redhat.com>
This forces the connection to abort if the application failed to
disconnect before flushing. This is aligned with how the common
flush services work.
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
There exists a race where the application can setup a connection
and then disconnect it before iw_cxgb4 processes the fw4_ack
message. For passive side connections, the fw4_ack message is
used to know when to stop the ep timer for MPA_REPLY messages.
If the application disconnects before the fw4_ack is handled then
c4iw_ep_disconnect() needs to clean up the timer state and stop the
timer before restarting it for the disconnect timer. Failure to do this
results in a "timer already started" message and a premature stopping
of the disconnect timer.
Fixes: e4b76a2 ("RDMA/iw_cxgb4: stop_ep_timer() after MPA negotiation")
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Elsewhere the sin_family field holds a value with a name of the form
AF_..., so it seems reasonable to do so here as well. Also the values
of PF_INET and AF_INET are the same.
The Coccinelle semantic patch that makes this change is as follows:
// <smpl>
@@
struct sockaddr_in sip;
@@
(
sip.sin_family ==
- PF_INET
+ AF_INET
|
sip.sin_family !=
- PF_INET
+ AF_INET
|
sip.sin_family =
- PF_INET
+ AF_INET
)
// </smpl>
Signed-off-by: Amitoj Kaur Chawla <amitoj1606@gmail.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
The commit 0f8ab0b6e9 ("RDMA/iw_cxgb4: Low resource fixes for Memory
registration") from Jun 10, 2016, leads to the following static checker
warning:
drivers/infiniband/hw/cxgb4/mem.c:612 c4iw_alloc_mw()
error: use kfree_skb() here instead of kfree(mhp->dereg_skb)
Also fixes skb leak in c4iw_dealloc_mw
Fixes: 0f8ab0b6e9 ("RDMA/iw_cxgb4: Low resource fixes for Memory registration")
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Steve Wise <swise@opengridcomputing.com>
Signed-off-by: Hariprasad Shenai <hariprasad@chelsio.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>
Fix sparse errors by making sure the fast assign destinations
are host cpu typed.
For the void __iomem *, just make the field match source
data.
Fix a bug where the hw_free trace printed the pointer vs.
the dereferenced value.
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@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>
The ftrace infrastructure used to evaluate the TRACE_SYSTEM
macro on every DEFINE_EVENT() macro. Now the TRACE_SYSTEM
macro only gets evaluated when trace/define_trace.h is
included, so the group event information is lost. This was
introduced in
commit acd388fd3a ("tracing: Give system name a pointer")
Therefore, each system tracepoint must be on its own file.
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@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>
