23 Mar, 2016
1 commit
-
Pull more rdma updates from Doug Ledford:
"Round two of 4.6 merge window patches.This is a monster pull request. I held off on the hfi1 driver updates
(the hfi1 driver is intimately tied to the qib driver and the new
rdmavt software library that was created to help both of them) in my
first pull request. The hfi1/qib/rdmavt update is probably 90% of
this pull request. The hfi1 driver is being left in staging so that
it can be fixed up in regards to the API that Al and yourself didn't
like. Intel has agreed to do the work, but in the meantime, this
clears out 300+ patches in the backlog queue and brings my tree and
their tree closer to sync.This also includes about 10 patches to the core and a few to mlx5 to
create an infrastructure for configuring SRIOV ports on IB devices.
That series includes one patch to the net core that we sent to netdev@
and Dave Miller with each of the three revisions to the series. We
didn't get any response to the patch, so we took that as implicit
approval.Finally, this series includes Intel's new iWARP driver for their x722
cards. It's not nearly the beast as the hfi1 driver. It also has a
linux-next merge issue, but that has been resolved and it now passes
just fine.Summary:
- A few minor core fixups needed for the next patch series
- The IB SRIOV series. This has bounced around for several versions.
Of note is the fact that the first patch in this series effects the
net core. It was directed to netdev and DaveM for each iteration
of the series (three versions total). Dave did not object, but did
not respond either. I've taken this as permission to move forward
with the series.- The new Intel X722 iWARP driver
- A huge set of updates to the Intel hfi1 driver. Of particular
interest here is that we have left the driver in staging since it
still has an API that people object to. Intel is working on a fix,
but getting these patches in now helps keep me sane as the upstream
and Intel's trees were over 300 patches apart"* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/dledford/rdma: (362 commits)
IB/ipoib: Allow mcast packets from other VFs
IB/mlx5: Implement callbacks for manipulating VFs
net/mlx5_core: Implement modify HCA vport command
net/mlx5_core: Add VF param when querying vport counter
IB/ipoib: Add ndo operations for configuring VFs
IB/core: Add interfaces to control VF attributes
IB/core: Support accessing SA in virtualized environment
IB/core: Add subnet prefix to port info
IB/mlx5: Fix decision on using MAD_IFC
net/core: Add support for configuring VF GUIDs
IB/{core, ulp} Support above 32 possible device capability flags
IB/core: Replace setting the zero values in ib_uverbs_ex_query_device
net/mlx5_core: Introduce offload arithmetic hardware capabilities
net/mlx5_core: Refactor device capability function
net/mlx5_core: Fix caching ATOMIC endian mode capability
ib_srpt: fix a WARN_ON() message
i40iw: Replace the obsolete crypto hash interface with shash
IB/hfi1: Add SDMA cache eviction algorithm
IB/hfi1: Switch to using the pin query function
IB/hfi1: Specify mm when releasing pages
...
22 Mar, 2016
5 commits
-
Following the practice exercised for network devices which allow the PF
net device to configure attributes of its virtual functions, we
introduce the following functions to be used by IPoIB which is the
network driver implementation for IB devices.ib_set_vf_link_state - set the policy for a VF link. More below.
ib_get_vf_config - read configuration information of a VF
ib_get_vf_stats - read VF statistics
ib_set_vf_guid - set the node or port GUID of a VFAlso add an indication in the device cap flags that indicates that this
IB devices is based on a virtual function.A VF shares the physical port with the PF and other VFs. When setting
the link state we have three options:1. Auto - in this mode, the virtual port follows the state of the
physical port and becomes active only if the physical port's state is
active. In all other cases it remains in a Down state.
2. Down - sets the state of the virtual port to Down
3. Up - causes the virtual port to transition into Initialize state if
it was not already in this state. A virtualization aware subnet manager
can then bring the state of the port into the Active state.Signed-off-by: Eli Cohen
Reviewed-by: Or Gerlitz
Signed-off-by: Doug Ledford -
Per the ongoing standardisation process, when virtual HCAs are present
in a network, traffic is routed based on a destination GID. In order to
access the SA we use the well known SA GID.We also add a GRH required boolean field to the port attributes which is
used to report to the verbs consumer whether this port is connected to a
virtual network. We use this field to realize whether we need to create
an address vector with GRH to access the subnet administrator. We clear
the port attributes struct before calling the hardware driver to make
sure the default remains that GRH is not required.Signed-off-by: Eli Cohen
Reviewed-by: Or Gerlitz
Signed-off-by: Doug Ledford -
The subnet prefix is a part of the port_info MAD returned and should be
available at the ib_port_attr struct. We define it here and provide a
default implementation in case the hardware driver does not provide one.
The subnet prefix is required when creating the address vector to access
the SA in networks where GRH must be used.Signed-off-by: Eli Cohen
Reviewed-by: Or Gerlitz
Signed-off-by: Doug Ledford -
The old bitwise device_cap_flags variable was limited to u32 which
has all bits already defined. In order to overcome it, we converted
device_cap_flags variable to be u64 type.Signed-off-by: Leon Romanovsky
Reviewed-by: Matan Barak
Signed-off-by: Doug Ledford
20 Mar, 2016
1 commit
-
Pull networking updates from David Miller:
"Highlights:1) Support more Realtek wireless chips, from Jes Sorenson.
2) New BPF types for per-cpu hash and arrap maps, from Alexei
Starovoitov.3) Make several TCP sysctls per-namespace, from Nikolay Borisov.
4) Allow the use of SO_REUSEPORT in order to do per-thread processing
of incoming TCP/UDP connections. The muxing can be done using a
BPF program which hashes the incoming packet. From Craig Gallek.5) Add a multiplexer for TCP streams, to provide a messaged based
interface. BPF programs can be used to determine the message
boundaries. From Tom Herbert.6) Add 802.1AE MACSEC support, from Sabrina Dubroca.
7) Avoid factorial complexity when taking down an inetdev interface
with lots of configured addresses. We were doing things like
traversing the entire address less for each address removed, and
flushing the entire netfilter conntrack table for every address as
well.8) Add and use SKB bulk free infrastructure, from Jesper Brouer.
9) Allow offloading u32 classifiers to hardware, and implement for
ixgbe, from John Fastabend.10) Allow configuring IRQ coalescing parameters on a per-queue basis,
from Kan Liang.11) Extend ethtool so that larger link mode masks can be supported.
From David Decotigny.12) Introduce devlink, which can be used to configure port link types
(ethernet vs Infiniband, etc.), port splitting, and switch device
level attributes as a whole. From Jiri Pirko.13) Hardware offload support for flower classifiers, from Amir Vadai.
14) Add "Local Checksum Offload". Basically, for a tunneled packet
the checksum of the outer header is 'constant' (because with the
checksum field filled into the inner protocol header, the payload
of the outer frame checksums to 'zero'), and we can take advantage
of that in various ways. From Edward Cree"* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-next: (1548 commits)
bonding: fix bond_get_stats()
net: bcmgenet: fix dma api length mismatch
net/mlx4_core: Fix backward compatibility on VFs
phy: mdio-thunder: Fix some Kconfig typos
lan78xx: add ndo_get_stats64
lan78xx: handle statistics counter rollover
RDS: TCP: Remove unused constant
RDS: TCP: Add sysctl tunables for sndbuf/rcvbuf on rds-tcp socket
net: smc911x: convert pxa dma to dmaengine
team: remove duplicate set of flag IFF_MULTICAST
bonding: remove duplicate set of flag IFF_MULTICAST
net: fix a comment typo
ethernet: micrel: fix some error codes
ip_tunnels, bpf: define IP_TUNNEL_OPTS_MAX and use it
bpf, dst: add and use dst_tclassid helper
bpf: make skb->tc_classid also readable
net: mvneta: bm: clarify dependencies
cls_bpf: reset class and reuse major in da
ldmvsw: Checkpatch sunvnet.c and sunvnet_common.c
ldmvsw: Add ldmvsw.c driver code
...
18 Mar, 2016
1 commit
-
Tracking user/QP ownership is needed to debug issues with
user ULPs like OpenMPI.Reviewed-by: Dennis Dalessandro
Signed-off-by: Mike Marciniszyn
Signed-off-by: Doug Ledford
17 Mar, 2016
3 commits
-
moved port mapper related code from drivers into common code
Signed-off-by: Mustafa Ismail
Signed-off-by: Tatyana E. Nikolova
Signed-off-by: Faisal Latif
Reviewed-by: Steve Wise
Tested-by: Steve Wise
Signed-off-by: Doug Ledford
15 Mar, 2016
1 commit
11 Mar, 2016
28 commits
-
The change requires a new pio_busy field in the iowait structure to
track the number of outstanding pios. The new counter together
with the sdma counter serve as the basis for a packet by packet decision
as to which egress mechanism to use. Since packets given to different
egress mechanisms are not ordered, this scheme will preserve the order.The iowait drain/wait mechanisms are extended for a pio case. An
additional qp wait flag is added for the PIO drain wait case.Currently the only pio wait is for buffers, so the no_bufs_available()
routine name is changed to pio_wait() and a third argument is passed
with one of the two pio wait flags to generalize the routine. A module
parameter is added to hold a configurable threshold. For now, the
module parameter is zero.A heuristic routine is added to return the func pointer of the proper
egress routine to use.The heuristic is as follows:
- SMI always uses pio
- GSI,UD qps threadhold use sdma
o No coordination with sdma is required because order is not required
and this qp pio count is not maintained for UD
- RC/UC ONLY packets threshold use SDMA
o If pio's are pending the pio_wait with the new wait flag is
called to delay for pios to drainThe threshold is potentially reduced by the QP's mtu.
The sc_buffer_alloc() has two additional args (a callback, a void *)
which are exploited by the RC/UC cases to pass a new complete routine
and a qp *.When the shadow ring completes the credit associated with a packet,
the new complete routine is called. The verbs_pio_complete() will then
decrement the busy count and trigger any drain waiters in qp destroy
or reset.Reviewed-by: Jubin John
Reviewed-by: Dennis Dalessandro
Signed-off-by: Mike Marciniszyn
Signed-off-by: Doug Ledford -
pahole noted the wasted 4 bytes after s_lock and r_lock.
Move s_flags and r_psn to fill the holes.
Reviewed-by: Dennis Dalessandro
Signed-off-by: Mike Marciniszyn
Signed-off-by: Doug Ledford -
Remove exported functions which are no longer required as the
functionality has moved into rdmavt. This also requires re-ordering some
of the functions since their prototype no longer appears in a header
file. Rather than add forward declarations it is just cleaner to
re-order some of the functions.Reviewed-by: Jubin John
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Initially it was intended that rdmavt would support some signaling
between the underlying driver and itself. However this turned out to be
unnecessary for qib and hfi1. If we need to add something like this in
later to support another driver we should do it then. As of now this
essentially dead code so remove it.Reviewed-by: Ira Weiny
Reviewed-by: Jubin John
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
While hfi1 and qib were still supporting bits and pieces of core verbs
components there needed to be a way to convey if rdmavt should handle
allocation and initialize of resources like the queue pair table. Now
that all of this is moved into rdmavt there is no need for these flags.
They are no longer used in the drivers.Reviewed-by: Ira Weiny
Reviewed-by: Jubin John
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Rdmavt adopted an smi_ah from qib which is not needed by hfi1. Move this
back to qib and get it out of the common library.Reviewed-by: Ira Weiny
Reviewed-by: Jubin John
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
For each verb validate that all requirements for driver callbacks are met.
If a function is called without checking for a valid pointer, it is a
required function. Also document what each callback function does.Reviewed-by: Mike Marciniszyn
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Add, remove, and otherwise clean up existing comments that are leftover
from the initial code postings of rdmavt. Many of the comments were added
to provide an idea on the direction we were thinking of going. Now that the
design is solidified make a pass over and clean everything up. Also add
details where lacking.Ensure all non static functions have nano comments.
Reviewed-by: Jubin John
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
This patch adds an additional lock to reduce contention on the s_lock.
This lock is used in post_send() so that the post_send is not
serialized with the send engine and other send related processing.To do this the s_next_psn is now maintained on post_send() while
post_send() related fields are moved to a new cache line. There is
an s_avail maintained for the post_send() to mitigate trading cache
lines with the send engine. The lock is released/acquired around
releasing the just built packet to the egress mechanism.Reviewed-by: Jubin John
Reviewed-by: Dennis Dalessandro
Signed-off-by: Dean Luick
Signed-off-by: Harish Chegondi
Signed-off-by: Mike Marciniszyn
Signed-off-by: Ira Weiny
Signed-off-by: Doug Ledford -
A busy_jiffies variable is maintained and updated when rc qps are
created and deleted. busy_jiffies is a scaled value of the number
of rc qps in the device. busy_jiffies is incremented every rc qp
scaling interval. busy_jiffies is added to the rc timeout
in add_retry_timer and mod_retry_timer. The rc qp scaling interval
is selected based on extensive performance evaluation of targeted
workloads.Reviewed-by: Dennis Dalessandro
Reviewed-by: Mike Marciniszyn
Signed-off-by: Vennila Megavannan
Signed-off-by: Jubin John
Signed-off-by: Doug Ledford -
The field is a vestige from ipath.
Reviewed-by: Dennis Dalessandro
Signed-off-by: Mike Marciniszyn
Signed-off-by: Doug Ledford -
LinkDownReason LocalMediaNotInstalled lacked an underscore
and was inconsistent with other defines in the same family.
This patch fixes this.Reviewed-by: Ira Weiny
Signed-off-by: Easwar Hariharan
Signed-off-by: Doug Ledford -
rvt_query_port calls into the driver through a call back function
query_port_state to populate the rest of ib_port_attr elements.
rvt_modify_port calls into the driver if needed through a call back
function shut_down_port()Signed-off-by: Harish Chegondi
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Addin query gid support. Rdmavt still relies on the driver to maintain
the gid table. Rdmavt simply calls into the driver to retrive the guid
for a particular port.Reviewed-by: Harish Chegondi
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
IB core uses 1 relative indexing for ports. All of our data structures
use 0 based indexing. Add an inline function that we can use whenever we
need to validate a legal value and try to convert a port number to a
port index at the entrance into rdmavt.Try to follow the policy that when we are talking about a port from IB
core point of view we refer to it as a port number. When port is an
index into our arrays refer to it as a port index.Reviewed-by: Mike Marciniszyn
Reviewed-by: Harish Chegondi
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Some hardware drivers requires additional checks on send WRs. Create an
optional call back to allow hardware drivers to reject a send WR.Reviewed-by: Dennis Dalessandro
Signed-off-by: Ira Weiny
Signed-off-by: Doug Ledford -
Fill in srq function stubs with code derived from hfi1 and qib.
Move necessary functions and data structure members as well.Reviewed-by: Dennis Dalessandro
Reviewed-by: Harish Chegondi
Signed-off-by: Jubin John
Signed-off-by: Doug Ledford -
Update all files added by rdmavt which do not yet have 2016 as the
copyright year.Reviewed-by: Ira Weiny
Reviewed-by: Harish Chegondi
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
This patch adds mad agent create and free to rdmavt.
Reviewed-by: Ira Weiny
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
This patch adds rdmavt device structure allocation in rdamvt. The
ib_device alloc is now done in rdmavt instead of the driver. Drivers
need to tell rdmavt the number of ports when calling.A side of effect of this patch is fixing a bug with port initialization
where the device structure port array was allocated over top of an
existing one.Reviewed-by: Ira Weiny
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Low level drivers need to be able to check incoming attributes as well as be
able to adjust their private data on queue pair modification. Add 2 driver
callbacks, check_modify_qp and modify_qp, to facilitate this.Signed-off-by: Ira Weiny
Signed-off-by: Doug Ledford -
s_sde should be in the low level driver QP private data.
Remove the definition from rvt_qp.
Signed-off-by: Ira Weiny
Signed-off-by: Doug Ledford -
This patch adds in the multicast add and remove functions as well as the
ancillary infrastructure needed.Reviewed-by: Mike Marciniszyn
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Add modify qp and supporting functions.
Reviewed-by: Mike Marciniszyn
Reviewed-by: Ira Weiny
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Add in a post_send and post_one_send to rdmavt. The ULP will provide a WQE
to rdmavt which will then walk and queue each element. Rdmavt will then
queue the work to be done in the driver or kick the driver's progress
routine.There needs to be a follow on patch which adds in another lock for the
head of the queue so that it can be added to and read from in parallel.
This will touch protocol handlers and require other changes in the
drivers. This will be done separately.Reviewed-by: Mike Marciniszyn
Reviewed-by: Ira Weiny
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
Brings in completion queue functionality. A kthread worker is added to
the rvt_dev_info to serve as a worker for completion queues.Reviewed-by: Mike Marciniszyn
Reviewed-by: Harish Chegondi
Signed-off-by: Dennis Dalessandro
Signed-off-by: Doug Ledford -
The current code is problematic when the QP creation and ipoib is
used to support NFS and NFS desires to do IO for paging purposes.
In that case, the GFP_KERNEL allocation within create_qp causes
a deadlock in tight memory situations.This fix adds support to create queue pair with GFP_NOIO flag for
connected mode only to cleanly fail the create queue pair in those
situations.This was previously fixed in qib but needed to get ported to hfi1.
This patch handles that for both hardwares in the new rdmavt common
layer.Reviewed-by: Dennis Dalessandro
Signed-off-by: Mike Marciniszyn
Signed-off-by: Doug Ledford -
With this commit, the drivers using rdmavt need not define query_device
function. But they should fill in the IB device attributes structure
rvt_dev_info.dparms.propsReviewed-by: Dennis Dalessandro
Signed-off-by: Harish Chegondi
Signed-off-by: Doug Ledford
