A list of MGID/MLID pairs is built when doing a multicast attach. When
the multicast detach is called, the list is searched, and regardless of
the search outcome, the driver detach is called.
If an MGID/MLID pair is not on the list, driver detach should not be
called, and an error should be returned. Calling the driver without
removing an MGID/MLID pair from the list can leave the core and driver
out of sync.
Linux commit:
20c7840a77ddcb2ed2fbd66e8197db2868495751
MFC after: 1 week
Sponsored by: Mellanox Technologies
When two handlers used the same object in the old schema, we blocked
the process in the kernel. The new schema just returns -EBUSY. This
could lead to different behaviour in applications between the old
schema and the new schema. In most cases, using such handlers
concurrently could lead to crashing the process. For example, if
thread A destroys a QP and thread B modifies it, we could have the
destruction happens before the modification. In this case, we are
accessing freed memory which could lead to crashing the process.
This is true for most cases. However, attaching and detaching
a multicast address from QP concurrently is safe. Therefore, we
preserve the original behaviour by adding a lock there.
Linux commit:
f48b726920d96dcd1860df06143bdea7d6d7dcc3
MFC after: 1 week
Sponsored by: Mellanox Technologies
When resolving an IP address in ibcore, only update the source address
upon normal completion. The ibcore address resolve function does not
care about the scope ID value of the IPv6 link-local addresses and expects
this information has already been extracted into the bound_dev_if field.
Because the same IPv6 link-local address can exist on multiple interfaces
the ibcore address resolver gets confused and returns ENETUNREACH.
Instead of updating both source address and bound_dev_if just keep the
address set to any address until resolving completes. For the sake of code
symmetry a similar change has been applied to the IPv4 address resolve path.
MFC after: 1 week
Sponsored by: Mellanox Technologies
When setting a large address resolve timeout it was observed that the
address resolving would succeed at the timeout and not when the address
was available. Make sure the address resolving requests are processed no
slower than one time every second.
While at it use "int" for jiffies instead of "unsigned long" to match
FreeBSD ticks.
MFC after: 1 week
Sponsored by: Mellanox Technologies
The original implementation used a reference to ifr_data and a cast to
do the equivalent of accessing ifr_addr. This was copied multiple
times since 1996.
Approved by: kib
MFC after: 1 week
Sponsored by: DARPA, AFRL
Differential Revision: https://reviews.freebsd.org/D14873
When exchanging CM messages the IPv6 scope ID should be ignored
for link local addresses when doing comparisons. Make sure the
scope ID is always set to zero for link local addresses.
MFC after: 1 week
Sponsored by: Mellanox Technologies
It is not enough to cancel delayed work structures before freeing.
Always cancel delayed work synchronously before freeing!
MFC after: 1 week
Sponsored by: Mellanox Technologies
Creating a UD address handle from user-space or from the kernel-space,
when the link layer is ethernet, requires resolving the remote L3
address into a L2 address. Doing this from the kernel is easy because
the required ARP(IPv4) and ND6(IPv6) address resolving APIs are readily
available. In userspace such an interface does not exist and kernel
help is required.
It should be noted that in an IP-based GID environment, the GID itself
does not contain all the information needed to resolve the destination
IP address. For example information like VLAN ID and SCOPE ID, is not
part of the GID and must be fetched from the GID attributes. Therefore
a source GID should always be referred to as a GID index. Instead of
going through various racy steps to obtain information about the
GID attributes from user-space, this is now all done by the kernel.
This patch optimises the L3 to L2 address resolving using the existing
create address handle uverbs interface, retrieving back the L2 address
as an additional user-space information structure.
This commit combines the following Linux upstream commits:
IB/core: Let create_ah return extended response to user
IB/core: Change ib_resolve_eth_dmac to use it in create AH
IB/mlx5: Make create/destroy_ah available to userspace
IB/mlx5: Use kernel driver to help userspace create ah
IB/mlx5: Report that device has udata response in create_ah
MFC after: 1 week
Sponsored by: Mellanox Technologies
This patch ensures the GID index is always used as a basis of resolving
incoming RDMA connections, as compared to the GID value itself.
Background:
On a per infiniband port basis, the GID identifier is not a unique identifier!
This assumption falls apart when VLAN ID, IPv6 scope ID and RoCE type,
as supported by RoCE v2, is taken into account. This additional
information is stored in the so-called GID attributes and is needed to
correctly identify the destination network interface for an incoming
connection.
Different VLANs are allowed to define the same IPv4 addresses and especially
for the default IPv6 link-local addresses or when using so-called containers
or jails, this is true.
The VNET information for the destination network interface is needed in
order to perform the L2 address lookup in the right Virtual Network Stack
context.
Consequently old functions previously used by RoCE v1, like
rdma_addr_find_smac_by_sgid() are impossible to support, because
there can be multiple identical GIDs associated with the same
infiniband port, and the answer to such a request becomes undefined.
This function has been removed.
MFC after: 1 week
Sponsored by: Mellanox Technologies
Implement the missing pieces in addr_resolve() to support loopback
addresses. IB core will test for the IFF_LOOPBACK flag in the network
interface and treat these devices in a special way.
MFC after: 1 week
Sponsored by: Mellanox Technologies
When a network device is departing, the RoCE GID entries should be
cleared before the default L2 link layer address is freed. Else a NULL
pointer access may happen.
MFC after: 1 week
Sponsored by: Mellanox Technologies
VLANs in ibcore.
IPv6 link local addresses are usually derived from the netdev MAC
address. This is applicable to VLAN devices and its lower netdevice as
well. In such cases the IPv6 link local address is a duplicate of the
default GID.
Now that link local IPv6 addresses based GIDs are supported, allow
adding such GID entries in the GID table.
MFC after: 1 week
Sponsored by: Mellanox Technologies
compilation under FreeBSD. The mthca driver was temporarily removed as
part of the Linux 4.9 RoCE/infinband upgrade.
Top commit in Linux source tree:
69973b830859bc6529a7a0468ba0d80ee5117826
Sponsored by: Mellanox Technologies
Mainly focus on files that use BSD 2-Clause license, however the tool I
was using misidentified many licenses so this was mostly a manual - error
prone - task.
The Software Package Data Exchange (SPDX) group provides a specification
to make it easier for automated tools to detect and summarize well known
opensource licenses. We are gradually adopting the specification, noting
that the tags are considered only advisory and do not, in any way,
superceed or replace the license texts.
No functional change intended.
Mainly focus on files that use BSD 3-Clause license.
The Software Package Data Exchange (SPDX) group provides a specification
to make it easier for automated tools to detect and summarize well known
opensource licenses. We are gradually adopting the specification, noting
that the tags are considered only advisory and do not, in any way,
superceed or replace the license texts.
Special thanks to Wind River for providing access to "The Duke of
Highlander" tool: an older (2014) run over FreeBSD tree was useful as a
starting point.
Different compilers may optimise the enum type in different ways. This ensures
coherency when range checking the value of enums in ibcore.
Sponsored by: Mellanox Technologies
MFC after: 1 week
variable. Access the variable directly instead of going through the sysctl()
interface in the kernel.
Sponsored by: Mellanox Technologies
MFC after: 1 week
Set the IPOIB_FLAG_INITIALIZED on dev_open and clear it on dev_stop to
avoid a race between ipoib load and the underlying device driver.
The device module must dispatch the IB_EVENT_PORT_ACTIVE event before ipoib
module is loaded. Otherwise, the flush will fail since no one set the
IPOIB_FLAG_INITIALIZED.
Submitted by: Slava Shwartsman <slavash@mellanox.com>
Sponsored by: Mellanox Technologies
MFC after: 1 week
Else the IPv6 address matching might fail. This change adds support for both
embedded and non-embedded IPv6 scope IDs when passing a IPv6 link-local socket
address to RDMA. Prior to this change only global IPv6 addresses would work
with RDMA.
Sponsored by: Mellanox Technologies
MFC after: 1 week
1) Fail to resolve RDMA address if rtalloc1() returns the loopback
device, lo0, as the gateway interface. Currently RDMA loopback is
not supported.
2) Use ip_dev_find() and ip6_dev_find() to lookup network interfaces
with matching IPv4 and IPv6 addresses, respectivly.
3) In addr_resolve() make sure the "ifa" pointer is always set, also when
the "ifp" is NULL. Else a NULL pointer access might happen trying to
read from the "ifa" pointer later on.
4) In rdma_addr_find_dmac_by_grh() make sure the "bound_dev_if" field
gets set properly instead of passing the scope ID through the IPv6
socket address structure. This is more in line with upstream OFED
in Linux.
5) In rdma_addr_find_smac_by_sgid() there is no need to pass the
scope ID for IPv6. Either it is stored in the "bound_dev_if" field
or ip6_dev_find() will find the correct network device regardless
of the scope ID.
Sponsored by: Mellanox Technologies
MFC after: 1 week
iWarp and RoCE in ibcore. The selection of RDMA_PS_TCP can not be used
to indicate iWarp protocol use. Backport the proper IB device
capabilities from Linux upstream to distinguish between iWarp and
RoCE. Only allocate the additional socket required for iWarp for RDMA
IDs when at least one iWarp device present. This resolves
interopability issues between iWarp and RoCE in ibcore
Reviewed by: np @
Differential Revision: https://reviews.freebsd.org/D12563
Sponsored by: Mellanox Technologies
MFC after: 3 days
- Add more sanity checks.
- Preserve source port number when resolving address.
- Remove no longer needed scope ID hacks for IPv6.
Sponsored by: Mellanox Technologies
This patch fixes an interopability issue between FreeBSD and non-FreeBSD
systems when the connection establishment is aborted. Refer to the
initial commit in Linux, drivers/infiniband/core/cm.c,
for a more detailed description.
Obtained from: Linux
MFC after: 3 days
Sponsored by: Mellanox Technologies
list.h includes a number of FreeBSD headers as a workaround for the
LIST_HEAD name collision. To reduce pollution, avoid including list.h
in commonly used headers when it is not explicitly needed.
Reviewed by: hselasky
MFC after: 1 week
Differential Revision: https://reviews.freebsd.org/D11249
This patch currently supports:
- ibcore as a kernel module only
- krping as a kernel module only
- ipoib as a kernel module only
Sponsored by: Mellanox Technologies
o Separate fields of struct socket that belong to listening from
fields that belong to normal dataflow, and unionize them. This
shrinks the structure a bit.
- Take out selinfo's from the socket buffers into the socket. The
first reason is to support braindamaged scenario when a socket is
added to kevent(2) and then listen(2) is cast on it. The second
reason is that there is future plan to make socket buffers pluggable,
so that for a dataflow socket a socket buffer can be changed, and
in this case we also want to keep same selinfos through the lifetime
of a socket.
- Remove struct struct so_accf. Since now listening stuff no longer
affects struct socket size, just move its fields into listening part
of the union.
- Provide sol_upcall field and enforce that so_upcall_set() may be called
only on a dataflow socket, which has buffers, and for listening sockets
provide solisten_upcall_set().
o Remove ACCEPT_LOCK() global.
- Add a mutex to socket, to be used instead of socket buffer lock to lock
fields of struct socket that don't belong to a socket buffer.
- Allow to acquire two socket locks, but the first one must belong to a
listening socket.
- Make soref()/sorele() to use atomic(9). This allows in some situations
to do soref() without owning socket lock. There is place for improvement
here, it is possible to make sorele() also to lock optionally.
- Most protocols aren't touched by this change, except UNIX local sockets.
See below for more information.
o Reduce copy-and-paste in kernel modules that accept connections from
listening sockets: provide function solisten_dequeue(), and use it in
the following modules: ctl(4), iscsi(4), ng_btsocket(4), ng_ksocket(4),
infiniband, rpc.
o UNIX local sockets.
- Removal of ACCEPT_LOCK() global uncovered several races in the UNIX
local sockets. Most races exist around spawning a new socket, when we
are connecting to a local listening socket. To cover them, we need to
hold locks on both PCBs when spawning a third one. This means holding
them across sonewconn(). This creates a LOR between pcb locks and
unp_list_lock.
- To fix the new LOR, abandon the global unp_list_lock in favor of global
unp_link_lock. Indeed, separating these two locks didn't provide us any
extra parralelism in the UNIX sockets.
- Now call into uipc_attach() may happen with unp_link_lock hold if, we
are accepting, or without unp_link_lock in case if we are just creating
a socket.
- Another problem in UNIX sockets is that uipc_close() basicly did nothing
for a listening socket. The vnode remained opened for connections. This
is fixed by removing vnode in uipc_close(). Maybe the right way would be
to do it for all sockets (not only listening), simply move the vnode
teardown from uipc_detach() to uipc_close()?
Sponsored by: Netflix
Differential Revision: https://reviews.freebsd.org/D9770
Support is implemented by mapping Linux's "struct net" into FreeBSD's
"struct vnet". Currently only vnet0 is supported by ibcore.
MFC after: 1 week
Sponsored by: Mellanox Technologies
Renumber cluase 4 to 3, per what everybody else did when BSD granted
them permission to remove clause 3. My insistance on keeping the same
numbering for legal reasons is too pedantic, so give up on that point.
Submitted by: Jan Schaumann <jschauma@stevens.edu>
Pull Request: https://github.com/freebsd/freebsd/pull/96
Sockets representing the TCP endpoints for iWARP connections are
allocated by the ibcore module. Before this revision they were closed
either by the ibcore module or the iw_cxgbe hardware driver depending on
the state transitions during connection teardown. This is error prone
and there were cases where both iw_cxgbe and ibcore closed the socket
leading to double-free panics. The fix is to let ibcore close the
sockets it creates and never do it in the driver.
- Use sodisconnect instead of soclose (preceded by solinger = 0) in the
driver to tear down an RDMA connection abruptly. This does what's
intended without releasing the socket's fd reference.
- Close the socket in ibcore when the iWARP iw_cm_id is destroyed. This
works for all kinds of sockets: clients that initiate connections,
listeners, and sockets accepted off of listeners.
Reviewed by: Steve Wise @ Open Grid Computing, hselasky@
MFC after: 3 days
Sponsored by: Chelsio Communications
Differential Revision: https://reviews.freebsd.org/D9796
like other PCI network drivers. The sys/ofed directory is now mainly
reserved for generic infiniband code, with exception of the mthca driver.
- Add new manual page, mlx4en(4), describing how to configure and load
mlx4en.
- All relevant driver C-files are now prefixed mlx4, mlx4_en and
mlx4_ib respectivly to avoid object filename collisions when compiling
the kernel. This also fixes an issue with proper dependency file
generation for the C-files in question.
- Device mlxen is now device mlx4en and depends on device mlx4, see
mlx4en(4). Only the network device name remains unchanged.
- The mlx4 and mlx4en modules are now built by default on i386 and
amd64 targets. Only building the mlx4ib module depends on
WITH_OFED=YES .
Sponsored by: Mellanox Technologies
by returning success on FIONBIO and FIOASYNC IOCTLs. The actual flags
handling is done by the kern_ioctl() function.
Reported by: Alex Bowden <alex.bowden@outlook.com>
Sponsored by: Mellanox Technologies
MFC after: 1 week
In pathological situations where the master subnet manager becomes
unresponsive for an extended period, we may otherwise end up queuing all
of the system's mbufs while waiting for a response to a path record lookup.
This addresses the same issue as commit 1e85b806f9 in Linux.
Reviewed by: cem, ngie
MFC after: 2 weeks
Sponsored by: EMC / Isilon Storage Division
This addresses a regression from an earlier upstream change which caused
cma_acquire_dev() to bypass the port GID cache and instead query the HCA
for each entry in its GID table. These queries can become extremely slow on
multiport devices, which has a negative impact on connection setup times.
Discussed with: hselasky
Obtained from: Linux
MFC after: 2 weeks
Sponsored by: EMC / Isilon Storage Division
SDP transmit and receive rings are always created in a sleepable context,
so we can use M_WAITOK and remove error checks.
Sponsored by: EMC / Isilon Storage Division
The generic socket may be detached from the PCB before the completion
queue is drained and destroyed, so this change closes a race condition
in connection teardown.
Sponsored by: EMC / Isilon Storage Division
tunable SYSCTL's. Linux module parameters are associated with the
module they belong to. FreeBSD does not share this concept of a parent
module. Instead add macros which define the prefix to use for the
module parameters in the LinuxKPI consumers.
While at it convert all "bool" LinuxKPI module parameters to "byte"
type, because we don't have a "bool" type of SYSCTL in FreeBSD.
Sponsored by: Mellanox Technologies
MFC after: 1 week
