Add a netevent callback for cma, mainly to catch NETEVENT_NEIGH_UPDATE.
Previously, when a system with failover MAC mechanism change its MAC address
during a CM connection attempt, the RDMA-CM would take a lot of time till
it disconnects and timesout due to the incorrect MAC address.
Now when we get a NETEVENT_NEIGH_UPDATE we check if it is due to a failover
MAC change and if so, we instantly destroy the CM and notify the user in order
to spare the unnecessary waiting for the timeout.
Link: https://lore.kernel.org/r/bb255c9e301cd50b905663b8e73f7f5133d0e4c5.1654601342.git.leonro@nvidia.com
Signed-off-by: Patrisious Haddad <phaddad@nvidia.com>
Reviewed-by: Mark Zhang <markzhang@nvidia.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Add to the cma, a tree that keeps track of all rdma_id_private channels
that were created while in RoCE mode.
The IDs are sorted first according to their netdevice ifindex then their
destination IP. And for IDs with matching IP they would be at the same node
in the tree, since the tree data is a list of all ids with matching destination IP.
The tree allows fast and efficient lookup of ids using an ifindex and
IP address which is useful for identifying relevant net_events promptly.
Link: https://lore.kernel.org/r/2fac52c86cc918c634ab24b3867d4aed992f54ec.1654601342.git.leonro@nvidia.com
Signed-off-by: Patrisious Haddad <phaddad@nvidia.com>
Reviewed-by: Mark Zhang <markzhang@nvidia.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Patchces for the merge window:
- Minor bug fixes in mlx5, mthca, pvrdma, rtrs, mlx4, hfi1, hns
- Minor cleanups: coding style, useless includes and documentation
- Reorganize how multicast processing works in rxe
- Replace a red/black tree with xarray in rxe which improves performance
- DSCP support and HW address handle re-use in irdma
- Simplify the mailbox command handling in hns
- Simplify iser now that FMR is eliminated
-----BEGIN PGP SIGNATURE-----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=aux1
-----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:
- Minor bug fixes in mlx5, mthca, pvrdma, rtrs, mlx4, hfi1, hns
- Minor cleanups: coding style, useless includes and documentation
- Reorganize how multicast processing works in rxe
- Replace a red/black tree with xarray in rxe which improves performance
- DSCP support and HW address handle re-use in irdma
- Simplify the mailbox command handling in hns
- Simplify iser now that FMR is eliminated
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (93 commits)
RDMA/nldev: Prevent underflow in nldev_stat_set_counter_dynamic_doit()
IB/iser: Fix error flow in case of registration failure
IB/iser: Generalize map/unmap dma tasks
IB/iser: Use iser_fr_desc as registration context
IB/iser: Remove iser_reg_data_sg helper function
RDMA/rxe: Use standard names for ref counting
RDMA/rxe: Replace red-black trees by xarrays
RDMA/rxe: Shorten pool names in rxe_pool.c
RDMA/rxe: Move max_elem into rxe_type_info
RDMA/rxe: Replace obj by elem in declaration
RDMA/rxe: Delete _locked() APIs for pool objects
RDMA/rxe: Reverse the sense of RXE_POOL_NO_ALLOC
RDMA/rxe: Replace mr by rkey in responder resources
RDMA/rxe: Fix ref error in rxe_av.c
RDMA/hns: Use the reserved loopback QPs to free MR before destroying MPT
RDMA/irdma: Add support for address handle re-use
RDMA/qib: Fix typos in comments
RDMA/mlx5: Fix memory leak in error flow for subscribe event routine
Revert "RDMA/core: Fix ib_qp_usecnt_dec() called when error"
RDMA/rxe: Remove useless argument for update_state()
...
If the state is not idle then resolve_prepare_src() should immediately
fail and no change to global state should happen. However, it
unconditionally overwrites the src_addr trying to build a temporary any
address.
For instance if the state is already RDMA_CM_LISTEN then this will corrupt
the src_addr and would cause the test in cma_cancel_operation():
if (cma_any_addr(cma_src_addr(id_priv)) && !id_priv->cma_dev)
Which would manifest as this trace from syzkaller:
BUG: KASAN: use-after-free in __list_add_valid+0x93/0xa0 lib/list_debug.c:26
Read of size 8 at addr ffff8881546491e0 by task syz-executor.1/32204
CPU: 1 PID: 32204 Comm: syz-executor.1 Not tainted 5.12.0-rc8-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
__dump_stack lib/dump_stack.c:79 [inline]
dump_stack+0x141/0x1d7 lib/dump_stack.c:120
print_address_description.constprop.0.cold+0x5b/0x2f8 mm/kasan/report.c:232
__kasan_report mm/kasan/report.c:399 [inline]
kasan_report.cold+0x7c/0xd8 mm/kasan/report.c:416
__list_add_valid+0x93/0xa0 lib/list_debug.c:26
__list_add include/linux/list.h:67 [inline]
list_add_tail include/linux/list.h:100 [inline]
cma_listen_on_all drivers/infiniband/core/cma.c:2557 [inline]
rdma_listen+0x787/0xe00 drivers/infiniband/core/cma.c:3751
ucma_listen+0x16a/0x210 drivers/infiniband/core/ucma.c:1102
ucma_write+0x259/0x350 drivers/infiniband/core/ucma.c:1732
vfs_write+0x28e/0xa30 fs/read_write.c:603
ksys_write+0x1ee/0x250 fs/read_write.c:658
do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46
entry_SYSCALL_64_after_hwframe+0x44/0xae
This is indicating that an rdma_id_private was destroyed without doing
cma_cancel_listens().
Instead of trying to re-use the src_addr memory to indirectly create an
any address derived from the dst build one explicitly on the stack and
bind to that as any other normal flow would do. rdma_bind_addr() will copy
it over the src_addr once it knows the state is valid.
This is similar to commit bc0bdc5afa ("RDMA/cma: Do not change
route.addr.src_addr.ss_family")
Link: https://lore.kernel.org/r/0-v2-e975c8fd9ef2+11e-syz_cma_srcaddr_jgg@nvidia.com
Cc: stable@vger.kernel.org
Fixes: 732d41c545 ("RDMA/cma: Make the locking for automatic state transition more clear")
Reported-by: syzbot+c94a3675a626f6333d74@syzkaller.appspotmail.com
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
XRC INI QPs should be able to adjust their local ACK timeout.
Fixes: 2c1619edef ("IB/cma: Define option to set ack timeout and pack tos_set")
Link: https://lore.kernel.org/r/1644421175-31943-1-git-send-email-haakon.bugge@oracle.com
Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com>
Suggested-by: Avneesh Pant <avneesh.pant@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
In RoCE we should use cma_iboe_set_mgid() and not cma_set_mgid to generate
the mgid, otherwise we will generate an IGMP for an incorrect address.
Fixes: b5de0c60cc ("RDMA/cma: Fix use after free race in roce multicast join")
Link: https://lore.kernel.org/r/913bc6783fd7a95fe71ad9454e01653ee6fb4a9a.1642491047.git.leonro@nvidia.com
Signed-off-by: Maor Gottlieb <maorg@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Currently, when cma_resolve_ib_dev() searches for a matching GID it will
stop searching after encountering the first empty GID table entry. This
behavior is wrong since neither IB nor RoCE spec enforce tightly packed
GID tables.
For example, when the matching valid GID entry exists at index N, and if a
GID entry is empty at index N-1, cma_resolve_ib_dev() will fail to find
the matching valid entry.
Fix it by making cma_resolve_ib_dev() continue searching even after
encountering missing entries.
Fixes: f17df3b0de ("RDMA/cma: Add support for AF_IB to rdma_resolve_addr()")
Link: https://lore.kernel.org/r/b7346307e3bb396c43d67d924348c6c496493991.1639055490.git.leonro@nvidia.com
Signed-off-by: Avihai Horon <avihaih@nvidia.com>
Reviewed-by: Mark Zhang <markzhang@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The existing tests are a little hard to comprehend. Use
check_add_overflow() instead.
Fixes: 04ded16724 ("RDMA/cma: Verify private data length")
Link: https://lore.kernel.org/r/1637661978-18770-1-git-send-email-haakon.bugge@oracle.com
Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Two list heads in the rdma_id_private are being used for multiple
purposes, to save a few bytes of memory. Give the different purposes
different names and union the memory that is clearly exclusive.
list splits into device_item and listen_any_item. device_item is threaded
onto the cma_device's list and listen_any goes onto the
listen_any_list. IDs doing any listen cannot have devices.
listen_list splits into listen_item and listen_list. listen_list is on the
parent listen any rdma_id_private and listen_item is on child listen that
is bound to a specific cma_dev.
Which name should be used in which case depends on the state and other
factors of the rdma_id_private. Remap all the confusing references to make
sense with the new names, so at least there is some hope of matching the
necessary preconditions with each access.
Link: https://lore.kernel.org/r/0-v1-a5ead4a0c19d+c3a-cma_list_head_jgg@nvidia.com
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The FSM can run in a circle allowing rdma_resolve_ip() to be called twice
on the same id_priv. While this cannot happen without going through the
work, it violates the invariant that the same address resolution
background request cannot be active twice.
CPU 1 CPU 2
rdma_resolve_addr():
RDMA_CM_IDLE -> RDMA_CM_ADDR_QUERY
rdma_resolve_ip(addr_handler) #1
process_one_req(): for #1
addr_handler():
RDMA_CM_ADDR_QUERY -> RDMA_CM_ADDR_BOUND
mutex_unlock(&id_priv->handler_mutex);
[.. handler still running ..]
rdma_resolve_addr():
RDMA_CM_ADDR_BOUND -> RDMA_CM_ADDR_QUERY
rdma_resolve_ip(addr_handler)
!! two requests are now on the req_list
rdma_destroy_id():
destroy_id_handler_unlock():
_destroy_id():
cma_cancel_operation():
rdma_addr_cancel()
// process_one_req() self removes it
spin_lock_bh(&lock);
cancel_delayed_work(&req->work);
if (!list_empty(&req->list)) == true
! rdma_addr_cancel() returns after process_on_req #1 is done
kfree(id_priv)
process_one_req(): for #2
addr_handler():
mutex_lock(&id_priv->handler_mutex);
!! Use after free on id_priv
rdma_addr_cancel() expects there to be one req on the list and only
cancels the first one. The self-removal behavior of the work only happens
after the handler has returned. This yields a situations where the
req_list can have two reqs for the same "handle" but rdma_addr_cancel()
only cancels the first one.
The second req remains active beyond rdma_destroy_id() and will
use-after-free id_priv once it inevitably triggers.
Fix this by remembering if the id_priv has called rdma_resolve_ip() and
always cancel before calling it again. This ensures the req_list never
gets more than one item in it and doesn't cost anything in the normal flow
that never uses this strange error path.
Link: https://lore.kernel.org/r/0-v1-3bc675b8006d+22-syz_cancel_uaf_jgg@nvidia.com
Cc: stable@vger.kernel.org
Fixes: e51060f08a ("IB: IP address based RDMA connection manager")
Reported-by: syzbot+dc3dfba010d7671e05f5@syzkaller.appspotmail.com
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
If the state is not idle then rdma_bind_addr() will immediately fail and
no change to global state should happen.
For instance if the state is already RDMA_CM_LISTEN then this will corrupt
the src_addr and would cause the test in cma_cancel_operation():
if (cma_any_addr(cma_src_addr(id_priv)) && !id_priv->cma_dev)
To view a mangled src_addr, eg with a IPv6 loopback address but an IPv4
family, failing the test.
This would manifest as this trace from syzkaller:
BUG: KASAN: use-after-free in __list_add_valid+0x93/0xa0 lib/list_debug.c:26
Read of size 8 at addr ffff8881546491e0 by task syz-executor.1/32204
CPU: 1 PID: 32204 Comm: syz-executor.1 Not tainted 5.12.0-rc8-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
__dump_stack lib/dump_stack.c:79 [inline]
dump_stack+0x141/0x1d7 lib/dump_stack.c:120
print_address_description.constprop.0.cold+0x5b/0x2f8 mm/kasan/report.c:232
__kasan_report mm/kasan/report.c:399 [inline]
kasan_report.cold+0x7c/0xd8 mm/kasan/report.c:416
__list_add_valid+0x93/0xa0 lib/list_debug.c:26
__list_add include/linux/list.h:67 [inline]
list_add_tail include/linux/list.h:100 [inline]
cma_listen_on_all drivers/infiniband/core/cma.c:2557 [inline]
rdma_listen+0x787/0xe00 drivers/infiniband/core/cma.c:3751
ucma_listen+0x16a/0x210 drivers/infiniband/core/ucma.c:1102
ucma_write+0x259/0x350 drivers/infiniband/core/ucma.c:1732
vfs_write+0x28e/0xa30 fs/read_write.c:603
ksys_write+0x1ee/0x250 fs/read_write.c:658
do_syscall_64+0x2d/0x70 arch/x86/entry/common.c:46
entry_SYSCALL_64_after_hwframe+0x44/0xae
Which is indicating that an rdma_id_private was destroyed without doing
cma_cancel_listens().
Instead of trying to re-use the src_addr memory to indirectly create an
any address build one explicitly on the stack and bind to that as any
other normal flow would do.
Link: https://lore.kernel.org/r/0-v1-9fbb33f5e201+2a-cma_listen_jgg@nvidia.com
Cc: stable@vger.kernel.org
Fixes: 732d41c545 ("RDMA/cma: Make the locking for automatic state transition more clear")
Reported-by: syzbot+6bb0528b13611047209c@syzkaller.appspotmail.com
Tested-by: Hao Sun <sunhao.th@gmail.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
If cma_listen_on_all() fails it leaves the per-device ID still on the
listen_list but the state is not set to RDMA_CM_ADDR_BOUND.
When the cmid is eventually destroyed cma_cancel_listens() is not called
due to the wrong state, however the per-device IDs are still holding the
refcount preventing the ID from being destroyed, thus deadlocking:
task:rping state:D stack: 0 pid:19605 ppid: 47036 flags:0x00000084
Call Trace:
__schedule+0x29a/0x780
? free_unref_page_commit+0x9b/0x110
schedule+0x3c/0xa0
schedule_timeout+0x215/0x2b0
? __flush_work+0x19e/0x1e0
wait_for_completion+0x8d/0xf0
_destroy_id+0x144/0x210 [rdma_cm]
ucma_close_id+0x2b/0x40 [rdma_ucm]
__destroy_id+0x93/0x2c0 [rdma_ucm]
? __xa_erase+0x4a/0xa0
ucma_destroy_id+0x9a/0x120 [rdma_ucm]
ucma_write+0xb8/0x130 [rdma_ucm]
vfs_write+0xb4/0x250
ksys_write+0xb5/0xd0
? syscall_trace_enter.isra.19+0x123/0x190
do_syscall_64+0x33/0x40
entry_SYSCALL_64_after_hwframe+0x44/0xa9
Ensure that cma_listen_on_all() atomically unwinds its action under the
lock during error.
Fixes: c80a0c52d8 ("RDMA/cma: Add missing error handling of listen_id")
Link: https://lore.kernel.org/r/20210913093344.17230-1-thomas.liu@ucloud.cn
Signed-off-by: Tao Liu <thomas.liu@ucloud.cn>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
ROCE uses IGMP for Multicast instead of the native Infiniband system where
joins are required in order to post messages on the Multicast group. On
Ethernet one can send Multicast messages to arbitrary addresses without
the need to subscribe to a group.
So ROCE correctly does not send IGMP joins during rdma_join_multicast().
F.e. in cma_iboe_join_multicast() we see:
if (addr->sa_family == AF_INET) {
if (gid_type == IB_GID_TYPE_ROCE_UDP_ENCAP) {
ib.rec.hop_limit = IPV6_DEFAULT_HOPLIMIT;
if (!send_only) {
err = cma_igmp_send(ndev, &ib.rec.mgid,
true);
}
}
} else {
So the IGMP join is suppressed as it is unnecessary.
However no such check is done in destroy_mc(). And therefore leaving a
sendonly multicast group will send an IGMP leave.
This means that the following scenario can lead to a multicast receiver
unexpectedly being unsubscribed from a MC group:
1. Sender thread does a sendonly join on MC group X. No IGMP join
is sent.
2. Receiver thread does a regular join on the same MC Group x.
IGMP join is sent and the receiver begins to get messages.
3. Sender thread terminates and destroys MC group X.
IGMP leave is sent and the receiver no longer receives data.
This patch adds the same logic for sendonly joins to destroy_mc() that is
also used in cma_iboe_join_multicast().
Fixes: ab15c95a17 ("IB/core: Support for CMA multicast join flags")
Link: https://lore.kernel.org/r/alpine.DEB.2.22.394.2109081340540.668072@gentwo.de
Signed-off-by: Christoph Lameter <cl@linux.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
From Maor Gottlieb
====================
Fix the use of nents and orig_nents in the sg table append helpers. The
nents should be used by the DMA layer to store the number of DMA mapped
sges, the orig_nents is the number of CPU sges.
Since the sg append logic doesn't always create a SGL with exactly
orig_nents entries store a total_nents as well to allow the table to be
properly free'd and reorganize the freeing logic to share across all the
use cases.
====================
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
* 'sg_nents':
RDMA: Use the sg_table directly and remove the opencoded version from umem
lib/scatterlist: Fix wrong update of orig_nents
lib/scatterlist: Provide a dedicated function to support table append
A MAD packet is sent as an unreliable datagram (UD). SA requests are sent
as MAD packets. As such, SA requests or responses may be silently dropped.
IB Core's MAD layer has a timeout and retry mechanism, which amongst
other, is used by RDMA CM. But it is not used by SA queries. The lack of
retries of SA queries leads to long specified timeout, and error being
returned in case of packet loss. The ULP or user-land process has to
perform the retry.
Fix this by taking advantage of the MAD layer's retry mechanism.
First, a check against a zero timeout is added in rdma_resolve_route(). In
send_mad(), we set the MAD layer timeout to one tenth of the specified
timeout and the number of retries to 10. The special case when timeout is
less than 10 is handled.
With this fix:
# ucmatose -c 1000 -S 1024 -C 1
runs stable on an Infiniband fabric. Without this fix, we see an
intermittent behavior and it errors out with:
cmatose: event: RDMA_CM_EVENT_ROUTE_ERROR, error: -110
(110 is ETIMEDOUT)
Link: https://lore.kernel.org/r/1628784755-28316-1-git-send-email-haakon.bugge@oracle.com
Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The net/sunrpc/xprtrdma module creates its QP using rdma_create_qp() and
immediately post receives, implicitly assuming the QP is in the INIT state
and thus valid for ib_post_recv().
The patch noted in Fixes: removed the RESET->INIT modifiy from
rdma_create_qp(), breaking NFS rdma for verbs providers that fail the
ib_post_recv() for a bad state.
This situation was proven using kprobes in rvt_post_recv() and
rvt_modify_qp(). The traces showed that the rvt_post_recv() failed before
ANY modify QP and that the current state was RESET.
Fix by reverting the patch below.
Fixes: dc70f7c3ed ("RDMA/cma: Remove unnecessary INIT->INIT transition")
Link: https://lore.kernel.org/r/1627583182-81330-1-git-send-email-mike.marciniszyn@cornelisnetworks.com
Cc: Haakon Bugge <haakon.bugge@oracle.com>
Cc: Chuck Lever III <chuck.lever@oracle.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Change location of rdma_restrack_del() to fix the bug where
task_struct was acquired but not released, causing to resource leak.
ucma_create_id() {
ucma_alloc_ctx();
rdma_create_user_id() {
rdma_restrack_new();
rdma_restrack_set_name() {
rdma_restrack_attach_task.part.0(); <--- task_struct was gotten
}
}
ucma_destroy_private_ctx() {
ucma_put_ctx();
rdma_destroy_id() {
_destroy_id() <--- id_priv was freed
}
}
}
Fixes: 889d916b6f ("RDMA/core: Don't access cm_id after its destruction")
Link: https://lore.kernel.org/r/073ec27acb943ca8b6961663c47c5abe78a5c8cc.1624948948.git.leonro@nvidia.com
Reported-by: Pavel Skripkin <paskripkin@gmail.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Fix a memory leak when "mda_resolve_route() is called more than once on
the same "rdma_cm_id".
This is possible if cma_query_handler() triggers the
RDMA_CM_EVENT_ROUTE_ERROR flow which puts the state machine back and
allows rdma_resolve_route() to be called again.
Link: https://lore.kernel.org/r/f6662b7b-bdb7-2706-1e12-47c61d3474b6@oracle.com
Signed-off-by: Gerd Rausch <gerd.rausch@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
An approximation for the PacketLifeTime is half the local ACK timeout.
The encoding for both timers are logarithmic.
If the local ACK timeout is set, but zero, it means the timer is
disabled. In this case, we choose the CMA_IBOE_PACKET_LIFETIME value,
since 50% of infinite makes no sense.
Before this commit, the PacketLifeTime became 255 if local ACK
timeout was zero (not running).
Fixed by explicitly testing for timeout being zero.
Fixes: e1ee1e62be ("RDMA/cma: Use ACK timeout for RoCE packetLifeTime")
Link: https://lore.kernel.org/r/1624371207-26710-1-git-send-email-haakon.bugge@oracle.com
Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The struct rdma_id_private contains three bit-fields, tos_set,
timeout_set, and min_rnr_timer_set. These are set by accessor functions
without any synchronization. If two or all accessor functions are invoked
in close proximity in time, there will be Read-Modify-Write from several
contexts to the same variable, and the result will be intermittent.
Fixed by protecting the bit-fields by the qp_mutex in the accessor
functions.
The consumer of timeout_set and min_rnr_timer_set is in
rdma_init_qp_attr(), which is called with qp_mutex held for connected
QPs. Explicit locking is added for the consumers of tos and tos_set.
This commit depends on ("RDMA/cma: Remove unnecessary INIT->INIT
transition"), since the call to rdma_init_qp_attr() from
cma_init_conn_qp() does not hold the qp_mutex.
Fixes: 2c1619edef ("IB/cma: Define option to set ack timeout and pack tos_set")
Fixes: 3aeffc46af ("IB/cma: Introduce rdma_set_min_rnr_timer()")
Link: https://lore.kernel.org/r/1624369197-24578-3-git-send-email-haakon.bugge@oracle.com
Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
In rdma_create_qp(), a connected QP will be transitioned to the INIT
state.
Afterwards, the QP will be transitioned to the RTR state by the
cma_modify_qp_rtr() function. But this function starts by performing an
ib_modify_qp() to the INIT state again, before another ib_modify_qp() is
performed to transition the QP to the RTR state.
Hence, there is no need to transition the QP to the INIT state in
rdma_create_qp().
Link: https://lore.kernel.org/r/1624369197-24578-2-git-send-email-haakon.bugge@oracle.com
Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
restrack should only be attached to a cm_id while the ID has a valid
device pointer. It is set up when the device is first loaded, but not
cleared when the device is removed. There is also two copies of the device
pointer, one private and one in the public API, and these were left out of
sync.
Make everything go to NULL together and manipulate restrack right around
the device assignments.
Found by syzcaller:
BUG: KASAN: wild-memory-access in __list_del include/linux/list.h:112 [inline]
BUG: KASAN: wild-memory-access in __list_del_entry include/linux/list.h:135 [inline]
BUG: KASAN: wild-memory-access in list_del include/linux/list.h:146 [inline]
BUG: KASAN: wild-memory-access in cma_cancel_listens drivers/infiniband/core/cma.c:1767 [inline]
BUG: KASAN: wild-memory-access in cma_cancel_operation drivers/infiniband/core/cma.c:1795 [inline]
BUG: KASAN: wild-memory-access in cma_cancel_operation+0x1f4/0x4b0 drivers/infiniband/core/cma.c:1783
Write of size 8 at addr dead000000000108 by task syz-executor716/334
CPU: 0 PID: 334 Comm: syz-executor716 Not tainted 5.11.0+ #271
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS
rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
Call Trace:
__dump_stack lib/dump_stack.c:79 [inline]
dump_stack+0xbe/0xf9 lib/dump_stack.c:120
__kasan_report mm/kasan/report.c:400 [inline]
kasan_report.cold+0x5f/0xd5 mm/kasan/report.c:413
__list_del include/linux/list.h:112 [inline]
__list_del_entry include/linux/list.h:135 [inline]
list_del include/linux/list.h:146 [inline]
cma_cancel_listens drivers/infiniband/core/cma.c:1767 [inline]
cma_cancel_operation drivers/infiniband/core/cma.c:1795 [inline]
cma_cancel_operation+0x1f4/0x4b0 drivers/infiniband/core/cma.c:1783
_destroy_id+0x29/0x460 drivers/infiniband/core/cma.c:1862
ucma_close_id+0x36/0x50 drivers/infiniband/core/ucma.c:185
ucma_destroy_private_ctx+0x58d/0x5b0 drivers/infiniband/core/ucma.c:576
ucma_close+0x91/0xd0 drivers/infiniband/core/ucma.c:1797
__fput+0x169/0x540 fs/file_table.c:280
task_work_run+0xb7/0x100 kernel/task_work.c:140
exit_task_work include/linux/task_work.h:30 [inline]
do_exit+0x7da/0x17f0 kernel/exit.c:825
do_group_exit+0x9e/0x190 kernel/exit.c:922
__do_sys_exit_group kernel/exit.c:933 [inline]
__se_sys_exit_group kernel/exit.c:931 [inline]
__x64_sys_exit_group+0x2d/0x30 kernel/exit.c:931
do_syscall_64+0x2d/0x40 arch/x86/entry/common.c:46
entry_SYSCALL_64_after_hwframe+0x44/0xa9
Fixes: 255d0c14b3 ("RDMA/cma: rdma_bind_addr() leaks a cma_dev reference count")
Link: https://lore.kernel.org/r/3352ee288fe34f2b44220457a29bfc0548686363.1620711734.git.leonro@nvidia.com
Signed-off-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The device attach triggers addition of CM_ID to the restrack DB.
However, when error occurs, we releasing this device, but defer CM_ID
release. This causes to the situation where restrack sees CM_ID that
is not valid anymore.
As a solution, add the CM_ID to the resource tracking DB only after the
attachment is finished.
Found by syzcaller:
infiniband syz0: added syz_tun
rdma_rxe: ignoring netdev event = 10 for syz_tun
infiniband syz0: set down
infiniband syz0: ib_query_port failed (-19)
restrack: ------------[ cut here ]------------
infiniband syz0: BUG: RESTRACK detected leak of resources
restrack: User CM_ID object allocated by syz-executor716 is not freed
restrack: ------------[ cut here ]------------
Fixes: b09c4d7012 ("RDMA/restrack: Improve readability in task name management")
Link: https://lore.kernel.org/r/ab93e56ba831eac65c322b3256796fa1589ec0bb.1618753862.git.leonro@nvidia.com
Signed-off-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
A switchdev RDMA device do not support IB CM. When such device is added to
the RDMA CM's device list, when application invokes rdma_listen(), cma
attempts to listen to such device, however it has IB CM attribute
disabled.
Due to this, rdma_listen() call fails to listen for other non switchdev
devices as well.
A below error message can be seen.
infiniband mlx5_0: RDMA CMA: cma_listen_on_dev, error -38
A failing call flow is below.
cma_listen_on_all()
cma_listen_on_dev()
_cma_attach_to_dev()
rdma_listen() <- fails on a specific switchdev device
This is because rdma_listen() is hardwired to only work with iwarp or IB
CM compatible devices.
Hence, when a IB device doesn't support IB CM or IW CM, avoid adding such
device to the cma list so rdma_listen() can't even be called.
Link: https://lore.kernel.org/r/f9cac00d52864ea7c61295e43fb64cf4db4fdae6.1618753862.git.leonro@nvidia.com
Signed-off-by: Parav Pandit <parav@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Introduce the ability for kernel ULPs to adjust the minimum RNR Retry
timer. The INIT -> RTR transition executed by RDMA CM will be used for
this adjustment. This avoids an additional ib_modify_qp() call.
rdma_set_min_rnr_timer() must be called before the call to rdma_connect()
on the active side and before the call to rdma_accept() on the passive
side.
The default value of RNR Retry timer is zero, which translates to 655
ms. When the receiver is not ready to accept a send messages, it encodes
the RNR Retry timer value in the NAK. The requestor will then wait at
least the specified time value before retrying the send.
The 5-bit value to be supplied to the rdma_set_min_rnr_timer() is
documented in IBTA Table 45: "Encoding for RNR NAK Timer Field".
Link: https://lore.kernel.org/r/1617216194-12890-2-git-send-email-haakon.bugge@oracle.com
Signed-off-by: Håkon Bugge <haakon.bugge@oracle.com>
Acked-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Do following cleanups about braces:
- Add the necessary braces to maintain context alignment.
- Fix the open '{' that is not on the same line as "switch".
- Remove braces that are not necessary for single statement blocks.
- Fix "else" that doesn't follow close brace '}'.
Link: https://lore.kernel.org/r/1617783353-48249-6-git-send-email-liweihang@huawei.com
Signed-off-by: Wenpeng Liang <liangwenpeng@huawei.com>
Signed-off-by: Weihang Li <liweihang@huawei.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Current code uses many different types when dealing with a port of a RDMA
device: u8, unsigned int and u32. Switch to u32 to clean up the logic.
This allows us to make (at least) the core view consistent and use the
same type. Unfortunately not all places can be converted. Many uverbs
functions expect port to be u8 so keep those places in order not to break
UAPIs. HW/Spec defined values must also not be changed.
With the switch to u32 we now can support devices with more than 255
ports. U32_MAX is reserved to make control logic a bit easier to deal
with. As a device with U32_MAX ports probably isn't going to happen any
time soon this seems like a non issue.
When a device with more than 255 ports is created uverbs will report the
RDMA device as having 255 ports as this is the max currently supported.
The verbs interface is not changed yet because the IBTA spec limits the
port size in too many places to be u8 and all applications that relies in
verbs won't be able to cope with this change. At this stage, we are
extending the interfaces that are using vendor channel solely
Once the limitation is lifted mlx5 in switchdev mode will be able to have
thousands of SFs created by the device. As the only instance of an RDMA
device that reports more than 255 ports will be a representor device and
it exposes itself as a RAW Ethernet only device CM/MAD/IPoIB and other
ULPs aren't effected by this change and their sysfs/interfaces that are
exposes to userspace can remain unchanged.
While here cleanup some alignment issues and remove unneeded sanity
checks (mainly in rdmavt),
Link: https://lore.kernel.org/r/20210301070420.439400-1-leon@kernel.org
Signed-off-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Commit ee1c60b1bf ("IB/SA: Modify SA to implicitly cache Class Port
info") removed the class_port_info_context struct usage, remove a couple
of leftovers.
Link: https://lore.kernel.org/r/20210314143427.76101-1-galpress@amazon.com
Signed-off-by: Gal Pressman <galpress@amazon.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Binding IPv6 address/port to AF_INET6 domain only is provided via
rdma_set_afonly(), but was not signalled to the provider. Applications
like NFS/RDMA bind the same port to both IPv4 and IPv6 addresses
simultaneously and thus rely on it working correctly.
Link: https://lore.kernel.org/r/20210219143441.1068-1-bmt@zurich.ibm.com
Tested-by: Chuck Lever <chuck.lever@oracle.com>
Tested-by: Benjamin Coddington <bcodding@redhat.com>
Signed-off-by: Bernard Metzler <bmt@zurich.ibm.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
ucma_process_join() allocates struct ucma_multicast mc and frees it if an
error occurs during its run. Specifically, if an error occurs in
copy_to_user(), a use-after-free might happen in the following scenario:
1. mc struct is allocated.
2. rdma_join_multicast() is called and succeeds. During its run,
cma_iboe_join_multicast() enqueues a work that will later use the
aforementioned mc struct.
3. copy_to_user() is called and fails.
4. mc struct is deallocated.
5. The work that was enqueued by cma_iboe_join_multicast() is run and
calls ucma_create_uevent() which tries to access mc struct (which is
freed by now).
Fix this bug by cancelling the work enqueued by cma_iboe_join_multicast().
Since cma_work_handler() frees struct cma_work, we don't use it in
cma_iboe_join_multicast() so we can safely cancel the work later.
The following syzkaller report revealed it:
BUG: KASAN: use-after-free in ucma_create_uevent+0x2dd/0x;3f0 drivers/infiniband/core/ucma.c:272
Read of size 8 at addr ffff88810b3ad110 by task kworker/u8:1/108
CPU: 1 PID: 108 Comm: kworker/u8:1 Not tainted 5.10.0-rc6+ #257
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
Workqueue: rdma_cm cma_work_handler
Call Trace:
__dump_stack lib/dump_stack.c:77 [inline]
dump_stack+0xbe/0xf9 lib/dump_stack.c:118
print_address_description.constprop.0+0x3e/0×60 mm/kasan/report.c:385
__kasan_report mm/kasan/report.c:545 [inline]
kasan_report.cold+0x1f/0×37 mm/kasan/report.c:562
ucma_create_uevent+0x2dd/0×3f0 drivers/infiniband/core/ucma.c:272
ucma_event_handler+0xb7/0×3c0 drivers/infiniband/core/ucma.c:349
cma_cm_event_handler+0x5d/0×1c0 drivers/infiniband/core/cma.c:1977
cma_work_handler+0xfa/0×190 drivers/infiniband/core/cma.c:2718
process_one_work+0x54c/0×930 kernel/workqueue.c:2272
worker_thread+0x82/0×830 kernel/workqueue.c:2418
kthread+0x1ca/0×220 kernel/kthread.c:292
ret_from_fork+0x1f/0×30 arch/x86/entry/entry_64.S:296
Allocated by task 359:
kasan_save_stack+0x1b/0×40 mm/kasan/common.c:48
kasan_set_track mm/kasan/common.c:56 [inline]
__kasan_kmalloc mm/kasan/common.c:461 [inline]
__kasan_kmalloc.constprop.0+0xc2/0xd0 mm/kasan/common.c:434
kmalloc include/linux/slab.h:552 [inline]
kzalloc include/linux/slab.h:664 [inline]
ucma_process_join+0x16e/0×3f0 drivers/infiniband/core/ucma.c:1453
ucma_join_multicast+0xda/0×140 drivers/infiniband/core/ucma.c:1538
ucma_write+0x1f7/0×280 drivers/infiniband/core/ucma.c:1724
vfs_write fs/read_write.c:603 [inline]
vfs_write+0x191/0×4c0 fs/read_write.c:585
ksys_write+0x1a1/0×1e0 fs/read_write.c:658
do_syscall_64+0x2d/0×40 arch/x86/entry/common.c:46
entry_SYSCALL_64_after_hwframe+0x44/0xa9
Freed by task 359:
kasan_save_stack+0x1b/0×40 mm/kasan/common.c:48
kasan_set_track+0x1c/0×30 mm/kasan/common.c:56
kasan_set_free_info+0x1b/0×30 mm/kasan/generic.c:355
__kasan_slab_free+0x112/0×160 mm/kasan/common.c:422
slab_free_hook mm/slub.c:1544 [inline]
slab_free_freelist_hook mm/slub.c:1577 [inline]
slab_free mm/slub.c:3142 [inline]
kfree+0xb3/0×3e0 mm/slub.c:4124
ucma_process_join+0x22d/0×3f0 drivers/infiniband/core/ucma.c:1497
ucma_join_multicast+0xda/0×140 drivers/infiniband/core/ucma.c:1538
ucma_write+0x1f7/0×280 drivers/infiniband/core/ucma.c:1724
vfs_write fs/read_write.c:603 [inline]
vfs_write+0x191/0×4c0 fs/read_write.c:585
ksys_write+0x1a1/0×1e0 fs/read_write.c:658
do_syscall_64+0x2d/0×40 arch/x86/entry/common.c:46
entry_SYSCALL_64_after_hwframe+0x44/0xa9
The buggy address belongs to the object at ffff88810b3ad100
which belongs to the cache kmalloc-192 of size 192
The buggy address is located 16 bytes inside of
192-byte region [ffff88810b3ad100, ffff88810b3ad1c0)
Fixes: b5de0c60cc ("RDMA/cma: Fix use after free race in roce multicast join")
Link: https://lore.kernel.org/r/20210211090517.1278415-1-leon@kernel.org
Reported-by: Amit Matityahu <mitm@nvidia.com>
Signed-off-by: Avihai Horon <avihaih@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
When a system receives a REREG event from the SM, then the SM information
in the kernel is marked as invalid and a request is sent to the SM to
update the information. The SM information is invalid in that time period.
However, receiving a REREG also occurs simultaneously in user space
applications that are now trying to rejoin the multicast groups. Some of
those may be sendonly multicast groups which are then failing.
If the SM information is invalid then ib_sa_sendonly_fullmem_support()
returns false. That is wrong because it just means that we do not know yet
if the potentially new SM supports sendonly joins.
Sendonly join was introduced in 2015 and all the Subnet managers have
supported it ever since. So there is no point in checking if a subnet
manager supports it.
Should an old opensm get a request for a sendonly join then the request
will fail. The code that is removed here accomodated that situation and
fell back to a full join.
Falling back to a full join is problematic in itself. The reason to use
the sendonly join was to reduce the traffic on the Infiniband fabric
otherwise one could have just stayed with the regular join. So this patch
may cause users of very old opensms to discover that lots of traffic
needlessly crosses their IB fabrics.
Link: https://lore.kernel.org/r/alpine.DEB.2.22.394.2101281845160.13303@www.lameter.com
Signed-off-by: Christoph Lameter <cl@linux.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
rdma_detroy_id() cannot be called under &lock - we must instead keep the
error'd ID around until &lock can be released, then destroy it.
This is complicated by the usual way listen IDs are destroyed through
cma_process_remove() which can run at any time and will asynchronously
destroy the same ID.
Remove the ID from visiblity of cma_process_remove() before going down the
destroy path outside the locking.
Fixes: c80a0c52d8 ("RDMA/cma: Add missing error handling of listen_id")
Link: https://lore.kernel.org/r/20201118133756.GK244516@ziepe.ca
Reported-by: syzbot+1bc48bf7f78253f664a9@syzkaller.appspotmail.com
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Don't silently continue if rdma_listen() fails but destroy previously
created CM_ID and return an error to the caller.
Fixes: d02d1f5359 ("RDMA/cma: Fix deadlock destroying listen requests")
Link: https://lore.kernel.org/r/20201104144008.3808124-5-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
There are two flows for handling RDMA_CM_EVENT_ROUTE_RESOLVED, either the
handler triggers a completion and another thread does rdma_connect() or
the handler directly calls rdma_connect().
In all cases rdma_connect() needs to hold the handler_mutex, but when
handler's are invoked this is already held by the core code. This causes
ULPs using the 2nd method to deadlock.
Provide a rdma_connect_locked() and have all ULPs call it from their
handlers.
Link: https://lore.kernel.org/r/0-v2-53c22d5c1405+33-rdma_connect_locking_jgg@nvidia.com
Reported-and-tested-by: Guoqing Jiang <guoqing.jiang@cloud.ionos.com>
Fixes: 2a7cec5381 ("RDMA/cma: Fix locking for the RDMA_CM_CONNECT state")
Acked-by: Santosh Shilimkar <santosh.shilimkar@oracle.com>
Acked-by: Jack Wang <jinpu.wang@cloud.ionos.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Max Gurtovoy <mgurtovoy@nvidia.com>
Reviewed-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The typical set of driver updates across the subsystem:
- Driver minor changes and bug fixes for mlx5, efa, rxe, vmw_pvrdma, hns,
usnic, qib, qedr, cxgb4, hns, bnxt_re
- Various rtrs fixes and updates
- Bug fix for mlx4 CM emulation for virtualization scenarios where MRA
wasn't working right
- Use tracepoints instead of pr_debug in the CM code
- Scrub the locking in ucma and cma to close more syzkaller bugs
- Use tasklet_setup in the subsystem
- Revert the idea that 'destroy' operations are not allowed to fail at
the driver level. This proved unworkable from a HW perspective.
- Revise how the umem API works so drivers make fewer mistakes using it
- XRC support for qedr
- Convert uverbs objects RWQ and MW to new the allocation scheme
- Large queue entry sizes for hns
- Use hmm_range_fault() for mlx5 On Demand Paging
- uverbs APIs to inspect the GID table instead of sysfs
- Move some of the RDMA code for building large page SGLs into
lib/scatterlist
-----BEGIN PGP SIGNATURE-----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=LKpl
-----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:
"A usual cycle for RDMA with a typical mix of driver and core subsystem
updates:
- Driver minor changes and bug fixes for mlx5, efa, rxe, vmw_pvrdma,
hns, usnic, qib, qedr, cxgb4, hns, bnxt_re
- Various rtrs fixes and updates
- Bug fix for mlx4 CM emulation for virtualization scenarios where
MRA wasn't working right
- Use tracepoints instead of pr_debug in the CM code
- Scrub the locking in ucma and cma to close more syzkaller bugs
- Use tasklet_setup in the subsystem
- Revert the idea that 'destroy' operations are not allowed to fail
at the driver level. This proved unworkable from a HW perspective.
- Revise how the umem API works so drivers make fewer mistakes using
it
- XRC support for qedr
- Convert uverbs objects RWQ and MW to new the allocation scheme
- Large queue entry sizes for hns
- Use hmm_range_fault() for mlx5 On Demand Paging
- uverbs APIs to inspect the GID table instead of sysfs
- Move some of the RDMA code for building large page SGLs into
lib/scatterlist"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (191 commits)
RDMA/ucma: Fix use after free in destroy id flow
RDMA/rxe: Handle skb_clone() failure in rxe_recv.c
RDMA/rxe: Move the definitions for rxe_av.network_type to uAPI
RDMA: Explicitly pass in the dma_device to ib_register_device
lib/scatterlist: Do not limit max_segment to PAGE_ALIGNED values
IB/mlx4: Convert rej_tmout radix-tree to XArray
RDMA/rxe: Fix bug rejecting all multicast packets
RDMA/rxe: Fix skb lifetime in rxe_rcv_mcast_pkt()
RDMA/rxe: Remove duplicate entries in struct rxe_mr
IB/hfi,rdmavt,qib,opa_vnic: Update MAINTAINERS
IB/rdmavt: Fix sizeof mismatch
MAINTAINERS: CISCO VIC LOW LATENCY NIC DRIVER
RDMA/bnxt_re: Fix sizeof mismatch for allocation of pbl_tbl.
RDMA/bnxt_re: Use rdma_umem_for_each_dma_block()
RDMA/umem: Move to allocate SG table from pages
lib/scatterlist: Add support in dynamic allocation of SG table from pages
tools/testing/scatterlist: Show errors in human readable form
tools/testing/scatterlist: Rejuvenate bit-rotten test
RDMA/ipoib: Set rtnl_link_ops for ipoib interfaces
RDMA/uverbs: Expose the new GID query API to user space
...
Separate IB_GID_TYPE_IB and IB_GID_TYPE_ROCE to two different values, so
enum ib_gid_type will match the gid types of the new query GID table API
which will be introduced in the following patches.
This change in enum ib_gid_type requires to change also enum
rdma_network_type by separating RDMA_NETWORK_IB and RDMA_NETWORK_ROCE_V1
values.
Link: https://lore.kernel.org/r/20200923165015.2491894-3-leon@kernel.org
Signed-off-by: Avihai Horon <avihaih@nvidia.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Functions related to nested interface infrastructure such as
netdev_walk_all_{ upper | lower }_dev() pass both private functions
and "data" pointer to handle their own things.
At this point, the data pointer type is void *.
In order to make it easier to expand common variables and functions,
this new netdev_nested_priv structure is added.
In the following patch, a new member variable will be added into this
struct to fix the lockdep issue.
Signed-off-by: Taehee Yoo <ap420073@gmail.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Use rdma_restrack_set_name() and rdma_restrack_parent_name() instead of
tricky uses of rdma_restrack_attach_task()/rdma_restrack_uadd().
This uniformly makes all restracks add'd using rdma_restrack_add().
Link: https://lore.kernel.org/r/20200922091106.2152715-6-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Have a single rdma_restrack_add() that adds an entry, there is no reason
to split the user/kernel here, the rdma_restrack_set_task() is responsible
for this difference.
This patch prepares the code to the future requirement of making restrack
is mandatory for managing ib objects.
Link: https://lore.kernel.org/r/20200922091106.2152715-5-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Refactor the restrack code to make sure the kref inside the restrack entry
properly kref's the object in which it is embedded. This slight change is
needed for future conversions of MR and QP which are refcounted before the
release and kfree.
The ideal flow from ib_core perspective as follows:
* Allocate ib_* structure with rdma_zalloc_*.
* Set everything that is known to ib_core to that newly created object.
* Initialize kref with restrack help
* Call to driver specific allocation functions.
* Insert into restrack DB
....
* Return and release restrack with restrack_put.
Largely this means a rdma_restrack_new() should be called near allocating
the containing structure.
Link: https://lore.kernel.org/r/20200922091106.2152715-4-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Update the code to have similar destroy pattern like other IB objects.
This change create asymmetry to the rdma_id_private create flow to make
sure that memory is managed by restrack.
Link: https://lore.kernel.org/r/20200922091106.2152715-2-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The roce path triggers a work queue that continues to touch the id_priv
but doesn't hold any reference on it. Futher, unlike in the IB case, the
work queue is not fenced during rdma_destroy_id().
This can trigger a use after free if a destroy is triggered in the
incredibly narrow window after the queue_work and the work starting and
obtaining the handler_mutex.
The only purpose of this work queue is to run the ULP event callback from
the standard context, so switch the design to use the existing
cma_work_handler() scheme. This simplifies quite a lot of the flow:
- Use the cma_work_handler() callback to launch the work for roce. This
requires generating the event synchronously inside the
rdma_join_multicast(), which in turn means the dummy struct
ib_sa_multicast can become a simple stack variable.
- cm_work_handler() used the id_priv kref, so we can entirely eliminate
the kref inside struct cma_multicast. Since the cma_multicast never
leaks into an unprotected work queue the kfree can be done at the same
time as for IB.
- Eliminating the general multicast.ib requires using cma_set_mgid() in a
few places to recompute the mgid.
Fixes: 3c86aa70bf ("RDMA/cm: Add RDMA CM support for IBoE devices")
Link: https://lore.kernel.org/r/20200902081122.745412-9-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Two places were open coding this sequence, and also pull in
cma_leave_roce_mc_group() which was called only once.
Link: https://lore.kernel.org/r/20200902081122.745412-8-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
There is no kernel user of RDMA CM multicast so this code managing the
multicast subscription of the kernel-only internal QP is dead. Remove it.
This makes the bug fixes in the next patches much simpler.
Link: https://lore.kernel.org/r/20200902081122.745412-7-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
These are the same thing, except that cma_ndev_work doesn't have a state
transition. Signal no state transition by setting old_state and new_state
== 0.
In all cases the handler function should not be called once
rdma_destroy_id() has progressed passed setting the state.
Link: https://lore.kernel.org/r/20200902081122.745412-6-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The only place that still uses it is rdma_join_multicast() which is only
doing a sanity check that the caller hasn't done something wrong and
doesn't need the spinlock.
At least in the case of rdma_join_multicast() the information it needs
will remain until the ID is destroyed once it enters these
states. Similarly there is no reason to check for these specific states in
the handler callback, instead use the usual check for a destroyed id under
the handler_mutex.
Link: https://lore.kernel.org/r/20200902081122.745412-5-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
There is a strange unlocked read of the ID state when checking for
reuseaddr. This is because an ID cannot be reusable once it becomes a
listening ID. Instead of using the state to exclude reuse, just clear it
as part of rdma_listen()'s flow to convert reusable into not reusable.
Once a ID goes to listen there is no way back out, and the only use of
reusable is on the bind_list check.
Finally, update the checks under handler_mutex to use READ_ONCE and audit
that once RDMA_CM_LISTEN is observed in a req callback it is stable under
the handler_mutex.
Link: https://lore.kernel.org/r/20200902081122.745412-4-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Re-organize things so the state variable is not read unlocked. The first
attempt to go directly from ADDR_BOUND immediately tells us if the ID is
already bound, if we can't do that then the attempt inside
rdma_bind_addr() to go from IDLE to ADDR_BOUND confirms the ID needs
binding.
Link: https://lore.kernel.org/r/20200902081122.745412-3-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Patrisious Haddad