There is a reference count error in error path code and a potential race
in check_rkey() in rxe_resp.c. When looking up the rkey for a memory
window the reference to the mw from rxe_lookup_mw() is dropped before a
reference is taken on the mr referenced by the mw. If the mr is destroyed
immediately after the call to rxe_put(mw) the mr pointer is unprotected
and may end up pointing at freed memory. The rxe_get(mr) call should take
place before the rxe_put(mw) call.
All errors in check_rkey() call rxe_put(mw) if mw is not NULL but it was
already called after the above. The mw pointer should be set to NULL after
the rxe_put(mw) call to prevent this from happening.
Fixes: cdd0b85675 ("RDMA/rxe: Implement memory access through MWs")
Link: https://lore.kernel.org/r/20230517211509.1819998-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
In rxe_net.c a received packet, from udp or loopback, is passed to
rxe_rcv() in rxe_recv.c as a udp packet. I.e. skb->data is pointing at the
udp header. But rxe_rcv() makes length checks to verify the packet is long
enough to hold the roce headers as if it were a roce
packet. I.e. skb->data pointing at the bth header. A runt packet would
appear to have 8 more bytes than it actually does which may lead to
incorrect behavior.
This patch calls skb_pull() to adjust the skb to point at the bth header
before calling rxe_rcv() which fixes this error.
Fixes: 8700e3e7c4 ("Soft RoCE driver")
Link: https://lore.kernel.org/r/20230517172242.1806340-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
rxe_cq_disable() has been removed but not its declaration.
Fixes: 78b26a3353 ("RDMA/rxe: Remove tasklet call from rxe_cq.c")
Link: https://lore.kernel.org/r/4f20ffc5-b2c4-0c11-2883-a835caf01a94@suse.com
Signed-off-by: Nicolas Morey <nmorey@suse.com>
Acked-by: Zhu Yanjun <zyjzyj2000@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
do_tcp_sendpages() is now just a small wrapper around tcp_sendmsg_locked(),
so inline it, allowing do_tcp_sendpages() to be removed. This is part of
replacing ->sendpage() with a call to sendmsg() with MSG_SPLICE_PAGES set.
Signed-off-by: David Howells <dhowells@redhat.com>
Reviewed-by: Bernard Metzler <bmt@zurich.ibm.com>
Reviewed-by: Tom Talpey <tom@talpey.com>
cc: Jason Gunthorpe <jgg@ziepe.ca>
cc: Leon Romanovsky <leon@kernel.org>
cc: Jens Axboe <axboe@kernel.dk>
cc: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The commit 9b4b7c1f9f ("RDMA/rxe: Add workqueue support for rxe tasks")
removed tasklets and replaced them with a workqueue, but relevant comments
are still remaining in the source code.
Fixes: 9b4b7c1f9f ("RDMA/rxe: Add workqueue support for rxe tasks")
Link: https://lore.kernel.org/r/20230518070027.942715-1-matsuda-daisuke@fujitsu.com
Signed-off-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com>
Acked-by: Zhu Yanjun <zyjzyj2000@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Replace tasklets by work queues for the three main rxe tasklets:
rxe_requester, rxe_completer and rxe_responder.
work queues are a more modern way to process work from an IRQ and provide
more control over how that work is run for future patches.
Link: https://lore.kernel.org/r/20230428171321.5774-1-rpearsonhpe@gmail.com
Signed-off-by: Ian Ziemba <ian.ziemba@hpe.com>
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Reviewed-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Tested-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Usual wide collection of unrelated items in drivers:
- Driver bug fixes and treewide cleanups in hfi1, siw, qib, mlx5, rxe,
usnic, usnic, bnxt_re, ocrdma, iser
* Unnecessary NULL checks
* kmap obsolescence
* pci_enable_pcie_error_reporting() obsolescence
* Unused variables and macros
* trace event related warnings
* casting warnings
- Code cleanups for irdm and erdma
- EFA reporting of 128 byte PCIe TLP support
- mlx5 more agressively uses the out of order HW feature
- Big rework of how state machines and tasks work in rxe
- Fix a syzkaller found crash netdev refcount leak in siw
- bnxt_re revises their HW description header
- Congestion control for bnxt_re
- Use mmu_notifiers more safely in hfi1
- mlx5 gets better support for PCIe relaxed ordering inside VMs
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQRRRCHOFoQz/8F5bUaFwuHvBreFYQUCZEva5wAKCRCFwuHvBreF
YZFmAQC9T3b/XQ3bRknYciuzbatC98o9xB0FTqmEFYGj+Y2lVAD9EEVe3HKfHfi3
t/GxXYB5r22oxg5bgsblZfEdEdTVCg8=
=akMm
-----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:
"Usual wide collection of unrelated items in drivers:
- Driver bug fixes and treewide cleanups in hfi1, siw, qib, mlx5,
rxe, usnic, usnic, bnxt_re, ocrdma, iser:
- remove unnecessary NULL checks
- kmap obsolescence
- pci_enable_pcie_error_reporting() obsolescence
- unused variables and macros
- trace event related warnings
- casting warnings
- Code cleanups for irdm and erdma
- EFA reporting of 128 byte PCIe TLP support
- mlx5 more agressively uses the out of order HW feature
- Big rework of how state machines and tasks work in rxe
- Fix a syzkaller found crash netdev refcount leak in siw
- bnxt_re revises their HW description header
- Congestion control for bnxt_re
- Use mmu_notifiers more safely in hfi1
- mlx5 gets better support for PCIe relaxed ordering inside VMs"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (81 commits)
RDMA/efa: Add rdma write capability to device caps
RDMA/mlx5: Use correct device num_ports when modify DC
RDMA/irdma: Drop spurious WQ_UNBOUND from alloc_ordered_workqueue() call
RDMA/rxe: Fix spinlock recursion deadlock on requester
RDMA/mlx5: Fix flow counter query via DEVX
RDMA/rxe: Protect QP state with qp->state_lock
RDMA/rxe: Move code to check if drained to subroutine
RDMA/rxe: Remove qp->req.state
RDMA/rxe: Remove qp->comp.state
RDMA/rxe: Remove qp->resp.state
RDMA/mlx5: Allow relaxed ordering read in VFs and VMs
net/mlx5: Update relaxed ordering read HCA capabilities
RDMA/mlx5: Check pcie_relaxed_ordering_enabled() in UMR
RDMA/mlx5: Remove pcie_relaxed_ordering_enabled() check for RO write
RDMA: Add ib_virt_dma_to_page()
RDMA/rxe: Fix the error "trying to register non-static key in rxe_cleanup_task"
RDMA/irdma: Slightly optimize irdma_form_ah_cm_frame()
RDMA/rxe: Fix incorrect TASKLET_STATE_SCHED check in rxe_task.c
IB/hfi1: Place struct mmu_rb_handler on cache line start
IB/hfi1: Fix bugs with non-PAGE_SIZE-end multi-iovec user SDMA requests
...
Every caller passes in zero, meaning they don't want any partial copy to
zero the remainder of the destination buffer.
Which is just as well, because the implementation of that function
didn't actually even look at that argument, and wasn't even aware it
existed, although some misleading comments did mention it still.
The 'zerorest' thing is a historical artifact of how "copy_from_user()"
worked, in that it would zero the rest of the kernel buffer that it
copied into.
That zeroing still exists, but it's long since been moved to generic
code, and the raw architecture-specific code doesn't do it. See
_copy_from_user() in lib/usercopy.c for this all.
However, while __copy_user_nocache() shares some history and superficial
other similarities with copy_from_user(), it is in many ways also very
different.
In particular, while the code makes it *look* similar to the generic
user copy functions that can copy both to and from user space, and take
faults on both reads and writes as a result, __copy_user_nocache() does
no such thing at all.
__copy_user_nocache() always copies to kernel space, and will never take
a page fault on the destination. What *can* happen, though, is that the
non-temporal stores take a machine check because one of the use cases is
for writing to stable memory, and any memory errors would then take
synchronous faults.
So __copy_user_nocache() does look a lot like copy_from_user(), but has
faulting behavior that is more akin to our old copy_in_user() (which no
longer exists, but copied from user space to user space and could fault
on both source and destination).
And it very much does not have the "zero the end of the destination
buffer", since a problem with the destination buffer is very possibly
the very source of the partial copy.
So this whole thing was just a confusing historical artifact from having
shared some code with a completely different function with completely
different use cases.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently the rxe driver makes little effort to make the changes to qp
state (which includes qp->attr.qp_state, qp->attr.sq_draining and
qp->valid) atomic between different client threads and IO threads. In
particular a common template is for an RDMA application to call
ib_modify_qp() to move a qp to ERR state and then wait until all the
packet and work queues have drained before calling ib_destroy_qp(). None
of these state changes are protected by locks to assure that the changes
are executed atomically and that memory barriers are included. This has
been observed to lead to incorrect behavior around qp cleanup.
This patch continues the work of the previous patches in this series and
adds locking code around qp state changes and lookups.
Link: https://lore.kernel.org/r/20230405042611.6467-5-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Move two blocks of code in rxe_comp.c and rxe_req.c to subroutines that
check if draining is complete in the SQD state and, if so, generate a
SQ_DRAINED event.
Link: https://lore.kernel.org/r/20230405042611.6467-4-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The rxe driver has four different QP state variables,
qp->attr.qp_state,
qp->req.state,
qp->comp.state, and
qp->resp.state.
All of these basically carry the same information.
This patch replaces uses of qp->req.state by qp->attr.qp_state and enum
rxe_qp_state. This is the third of three patches which will remove all
but the qp->attr.qp_state variable. This will bring the driver closer to
the IBA description.
Link: https://lore.kernel.org/r/20230405042611.6467-3-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The rxe driver has four different QP state variables,
qp->attr.qp_state,
qp->req.state,
qp->comp.state, and
qp->resp.state.
All of these basically carry the same information.
This patch replaces uses of qp->comp.state by qp->attr.qp_state. This is
the second of three patches which will remove all but the
qp->attr.qp_state variable. This will bring the driver closer to the IBA
description.
Link: https://lore.kernel.org/r/20230405042611.6467-2-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The rxe driver has four different QP state variables,
qp->attr.qp_state,
qp->req.state,
qp->comp.state, and
qp->resp.state.
All of these basically carry the same information.
This patch replaces uses of qp->resp.state by qp->attr.qp_state. This is
the first of three patches which will remove all but the qp->attr.qp_state
variable. This will bring the driver closer to the IBA description.
Link: https://lore.kernel.org/r/20230405042611.6467-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Make it clearer what is going on by adding a function to go back from the
"virtual" dma_addr to a kva and another to a struct page. This is used in the
ib_uses_virt_dma() style drivers (siw, rxe, hfi, qib).
Call them instead of a naked casting and virt_to_page() when working with dma_addr
values encoded by the various ib_map functions.
This also fixes the virt_to_page() casting problem Linus Walleij has been
chasing.
Cc: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Link: https://lore.kernel.org/r/0-v2-05ea785520ed+10-ib_virt_page_jgg@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
syzbot is reporting that siw_netdev_event(NETDEV_UNREGISTER) cannot destroy
siw_device created after unshare(CLONE_NEWNET) due to net namespace check.
It seems that this check was by error there and should be removed.
Reported-by: syzbot <syzbot+5e70d01ee8985ae62a3b@syzkaller.appspotmail.com>
Link: https://syzkaller.appspot.com/bug?extid=5e70d01ee8985ae62a3b
Suggested-by: Jason Gunthorpe <jgg@ziepe.ca>
Suggested-by: Leon Romanovsky <leon@kernel.org>
Fixes: bdcf26bf9b ("rdma/siw: network and RDMA core interface")
Signed-off-by: Tetsuo Handa <penguin-kernel@I-love.SAKURA.ne.jp>
Link: https://lore.kernel.org/r/a44e9ac5-44e2-d575-9e30-02483cc7ffd1@I-love.SAKURA.ne.jp
Reviewed-by: Bernard Metzler <bmt@zurich.ibm.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Remove the tasklet call in rxe_cq.c and also the is_dying in the
cq struct. There is no reason for the rxe driver to defer the call
to the cq completion handler by scheduling a tasklet. rxe_cq_post()
is not called in a hard irq context.
The rxe driver currently is incorrect because the tasklet call is
made without protecting the cq pointer with a reference from having
the underlying memory freed before the deferred routine is called.
Executing the comp_handler inline fixes this problem.
Fixes: 8700e3e7c4 ("Soft RoCE driver")
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Link: https://lore.kernel.org/r/20230327215643.10410-1-rpearsonhpe@gmail.com
Acked-by: Zhu Yanjun <zyjzyj2000@gmail.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
This patch is a major rewrite of the tasklet routines in rxe_task.c. The
main motivation for this is the realization that the code violates the
safety of the qp pointer by correct reference counting. When a tasklet is
scheduled from a verbs API the calling thread has a valid reference to the
qp and schedules the tasklet to run at a later time carrying a pointer to
the qp. Once the calling code returns however the qp can be destroyed at
any time. In order to correct this a reference to the qp must be taken
when the task is scheduled and held until it finishes running. This is
complicated by the tasklet library not alwys running a task that is
scheduled depending on whether someone else has scheduled it.
This patch moves the logic for deciding whether to run or schedule a task
outside of do_task() and guarantees that there is only one copy of the
task scheduled or running at a time.
Secondly the separate flags controlling teardown and draining of the task
are included in the task state machine and all references to the state are
protected by spinlocks to avoid consistency and memory barrier issues.
Link: https://lore.kernel.org/r/20230304174533.11296-9-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Replace rxe_run_task() by rxe_sched_task() when tasks call each other.
These are not performance critical and mainly involve error paths but they
run the risk of causing deadlocks.
Link: https://lore.kernel.org/r/20230304174533.11296-8-rpearsonhpe@gmail.com
Signed-off-by: Ian Ziemba <ian.ziemba@hpe.com>
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The subroutine __rxe_do_task is not thread safe and it has no way to
guarantee that the tasks, which are designed with the assumption that they
are non-reentrant, are not reentered. All of its uses are non-performance
critical.
This patch replaces calls to __rxe_do_task with calls to
rxe_sched_task. It also removes irrelevant or unneeded if tests.
Instead of calling the task machinery a single call to the tasklet
function (rxe_requester, etc.) is sufficient to draing the queues if task
execution has been disabled or stopped.
Together these changes allow the removal of __rxe_do_task.
Link: https://lore.kernel.org/r/20230304174533.11296-7-rpearsonhpe@gmail.com
Signed-off-by: Ian Ziemba <ian.ziemba@hpe.com>
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Currently each of the three tasklets requester, completer and responder in
the rxe driver take and release a reference to the qp argument at the
beginning and end of the subroutines. The caller passing in the qp
argument should be responsible for holding a reference to qp so these are
not required. Further doing so breaks the qp cleanup code in
rxe_qp_do_cleanup which calls these routines after all the references have
been dropped so they cannot drain the packet and work request queues as
intended.
In fact if these routines are deferred by calling tasklet_schedule there
is no guarantee that the calling code does have a qp reference. That is a
bug in rxe_task.c which will be fixed later in this series.
Link: https://lore.kernel.org/r/20230304174533.11296-6-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Cleanup the handling of qp in the error state, reset state and during
rxe_qp_do_cleanup. Make the same as rxe_resp.c
Link: https://lore.kernel.org/r/20230304174533.11296-5-rpearsonhpe@gmail.com
Signed-off-by: Ian Ziemba <ian.ziemba@hpe.com>
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Cleanup the handling of qp in the error state, reset state and during
rxe_qp_do_cleanup. The error state does about the same thing as the others
but has code spread all over.
This patch combines them in a cleaner way.
Link: https://lore.kernel.org/r/20230304174533.11296-4-rpearsonhpe@gmail.com
Signed-off-by: Ian Ziemba <ian.ziemba@hpe.com>
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Originally is was thought that the tasklet machinery in rxe_task.c would
be used in other applications but that has not happened for years. This
patch replaces the 'void *arg' by struct 'rxe_qp *qp' in the parameters to
the tasklet calls. This change will have no affect on performance but may
make the code a little clearer.
Link: https://lore.kernel.org/r/20230304174533.11296-2-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
This patch adds error and debug messages so that every interaction
with rdma-core through a verbs API call or a completion error return
will generate at least one error message backed up by debug messages
with more detail.
With dynamic debugging one can follow up after seeing an error message
by turning on the appropriate debug messages.
Link: https://lore.kernel.org/r/20230303221623.8053-5-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Extend the dbg log messages (e.g. rxe_dbg_xxx) to include
err and info types. rxe.c is modified to use these new log
messages as examples.
Link: https://lore.kernel.org/r/20230303221623.8053-4-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Replace the name rxe_dbg with rxe_dbg_dev which better matches
the remaining rxe_dbg_xxx macros for debug messages with a
rxe device parameter. Reuse the name rxe_dbg for debug messages
which do not have a rxe device parameter.
Link: https://lore.kernel.org/r/20230303221623.8053-3-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
'exists' looks like a boolean. This patch replaces it by the
normal name used for the rxe device, 'rxe', which should be a
little less confusing. The second rxe_dbg() message is
incorrect since rxe is known to be NULL and this will cause a
seg fault if this message were ever sent. Replace it by pr_debug
for the moment.
Fixes: c6aba5ea00 ("RDMA/rxe: Replace pr_xxx by rxe_dbg_xxx in rxe.c")
Link: https://lore.kernel.org/r/20230303221623.8053-2-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
There is no need to check 'rdi->qp_dev' for NULL. The field 'qp_dev'
is created in rvt_register_device() which will fail if the 'qp_dev'
allocation fails in rvt_driver_qp_init(). Overwise this pointer
doesn't changed and passed to rvt_qp_exit() by the next step.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: 0acb0cc7ec ("IB/rdmavt: Initialize and teardown of qpn table")
Signed-off-by: Natalia Petrova <n.petrova@fintech.ru>
Link: https://lore.kernel.org/r/20230303124408.16685-1-n.petrova@fintech.ru
Signed-off-by: Leon Romanovsky <leon@kernel.org>
The "cplen" result used by the memcpy() into struct rvt_swqe "wqe" may
be sized to 80 for struct rvt_ud_wr (which is member "ud_wr", not "wr"
which is only 40 bytes in size). Change the destination union member so
the compiler can use the correct bounds check.
struct rvt_swqe {
union {
struct ib_send_wr wr; /* don't use wr.sg_list */
struct rvt_ud_wr ud_wr;
...
};
...
};
Silences false positive memcpy() run-time warning:
memcpy: detected field-spanning write (size 80) of single field "&wqe->wr" at drivers/infiniband/sw/rdmavt/qp.c:2043 (size 40)
Reported-by: Zhang Yi <yizhan@redhat.com>
Link: https://bugzilla.kernel.org/show_bug.cgi?id=216561
Cc: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Leon Romanovsky <leon@kernel.org>
Cc: linux-rdma@vger.kernel.org
Signed-off-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20230218185701.never.779-kees@kernel.org
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
When seg is equal to MAX_ARRAY, the loop should break, otherwise
it will result in out of range access.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: b9be6f18cf ("rdma/siw: transmit path")
Signed-off-by: Daniil Dulov <d.dulov@aladdin.ru>
Link: https://lore.kernel.org/r/20230227091751.589612-1-d.dulov@aladdin.ru
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Small cycle this time:
- Minor driver updates for hfi1, cxgb4, erdma, hns, irdma, mlx5, siw, mana
- inline CQE support for hns
- Have mlx5 display device error codes
- Pinned DMABUF support for irdma
- Continued rxe cleanups, particularly converting the MRs to use xarray
- Improvements to what can be cached in the mlx5 mkey cache
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQRRRCHOFoQz/8F5bUaFwuHvBreFYQUCY/gPmgAKCRCFwuHvBreF
YW5IAP4xOAiTif4f87vD1twRU/ebq4VEX0r+C2NX5x5fwlCJrAEA7RLV8uG9Uii2
ez0BuWNxfajuvFHntnZ1E+7UDP0S8gk=
=CgUH
-----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:
"Quite a small cycle this time, even with the rc8. I suppose everyone
went to sleep over xmas.
- Minor driver updates for hfi1, cxgb4, erdma, hns, irdma, mlx5, siw,
mana
- inline CQE support for hns
- Have mlx5 display device error codes
- Pinned DMABUF support for irdma
- Continued rxe cleanups, particularly converting the MRs to use
xarray
- Improvements to what can be cached in the mlx5 mkey cache"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (61 commits)
IB/mlx5: Extend debug control for CC parameters
IB/hfi1: Fix sdma.h tx->num_descs off-by-one errors
IB/hfi1: Fix math bugs in hfi1_can_pin_pages()
RDMA/irdma: Add support for dmabuf pin memory regions
RDMA/mlx5: Use query_special_contexts for mkeys
net/mlx5e: Use query_special_contexts for mkeys
net/mlx5: Change define name for 0x100 lkey value
net/mlx5: Expose bits for querying special mkeys
RDMA/rxe: Fix missing memory barriers in rxe_queue.h
RDMA/mana_ib: Fix a bug when the PF indicates more entries for registering memory on first packet
RDMA/rxe: Remove rxe_alloc()
RDMA/cma: Distinguish between sockaddr_in and sockaddr_in6 by size
Subject: RDMA/rxe: Handle zero length rdma
iw_cxgb4: Fix potential NULL dereference in c4iw_fill_res_cm_id_entry()
RDMA/mlx5: Use rdma_umem_for_each_dma_block()
RDMA/umem: Remove unused 'work' member from struct ib_umem
RDMA/irdma: Cap MSIX used to online CPUs + 1
RDMA/mlx5: Check reg_create() create for errors
RDMA/restrack: Correct spelling
RDMA/cxgb4: Fix potential null-ptr-deref in pass_establish()
...
An earlier patch which introduced smp_load_acquire/smp_store_release
into rxe_queue.h incorrectly assumed that surrounding spin-locks in
rxe_verbs.c around queue updates for kernel ulps was sufficient to
protect the passing of data through the queues between the ulp and
the rxe tasklets. But this was incorrect. The typical sequence was
ulp rxe requester tasklet
------------------------ ---------------------
spin_lock_irqsave() wqe = queue_head(queue)
if (!queue_full(q)) { if (!wqe)
spin_unlock_irqrestore return;
return -ENOMEM
} <process wqe>
wqe = queue_producer_addr(q)
<fill in wqe> queue_advance_consumer(queue)
queue_advance_producer(q)
spin_unlock_irqrestore()
queue_head() calls queue_empty() which calls smp_load_acquire()
For user space apps queue_advance_producer() calls smp_store_release()
so that there is a memory barrier between the producer and the
consumer but for kernel ulps queue_advance_produce() just incremented
the producer index because the lock function is a release function.
But to work the barrier has to come between filling in the wqe and
updating the producer index. This patch adds the missing barriers.
It also changes the enum names for the ulp queue types to
QUEUE_TYPE_FROM/TO_ULP instead of QUEUE_TYPE_TO/FROM_DRIVER
which is very ambiguous. This bug is suspected as the cause of very
rare lockups in a very high scale storage application. It is a bug
in any case and should be corrected.
Fixes: 0a67c46d2e ("RDMA/rxe: Protect user space index loads/stores")
Link: https://lore.kernel.org/r/20230214071053.5395-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Currently all the object types in the rxe driver are allocated in
rdma-core except for MRs. By moving tha kzalloc() call outside of
the pool code the rxe_alloc() subroutine can be eliminated and code
checking for MR as a special case can be removed.
This patch moves the kzalloc() and kfree_rcu() calls into the mr
registration and destruction verbs. It removes that code from
rxe_pool.c including the rxe_alloc() subroutine which is no longer
used.
Link: https://lore.kernel.org/r/20230213225551.12437-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Reviewed-by: Devesh Sharma <devesh.s.sharma@oracle.com>
Reviewed-by: Devesh Sharma <devesh.s.sharma@oracle.com>
Reviewed-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Currently the rxe driver does not handle all cases of zero length rdma
operations correctly. The client does not have to provide an rkey for zero
length RDMA read or write operations so the rkey provided may be invalid
and should not be used to lookup an mr.
This patch corrects the driver to ignore the provided rkey if the reth
length is zero for read or write operations and make sure to set the mr to
NULL. In read_reply() if length is zero rxe_recheck_mr() is not
called. Warnings are added in the routines in rxe_mr.c to catch NULL MRs
when the length is non-zero.
Fixes: 8700e3e7c4 ("Soft RoCE driver")
Link: https://lore.kernel.org/r/20230202044240.6304-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Reviewed-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
To avoid racing with other user memory reservations, immediately
account full amount of pages to be pinned.
Fixes: 2251334dca ("rdma/siw: application buffer management")
Reported-by: Jason Gunthorpe <jgg@nvidia.com>
Suggested-by: Alistair Popple <apopple@nvidia.com>
Reviewed-by: Alistair Popple <apopple@nvidia.com>
Signed-off-by: Bernard Metzler <bmt@zurich.ibm.com>
Link: https://lore.kernel.org/r/20230202101000.402990-1-bmt@zurich.ibm.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Replace struct rxe-phys_buf and struct rxe_map by struct xarray
in rxe_verbs.h. This allows using rcu locking on reads for
the memory maps stored in each mr.
This is based off of a sketch of a patch from Jason Gunthorpe in the
link below. Some changes were needed to make this work. It applies
cleanly to the current for-next and passes the pyverbs, perftest
and the same blktests test cases which run today.
Link: https://lore.kernel.org/r/20230119235936.19728-7-rpearsonhpe@gmail.com
Link: https://lore.kernel.org/linux-rdma/Y3gvZr6%2FNCii9Avy@nvidia.com/
Co-developed-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Cleanup usage of mr->page_shift and mr->page_mask and introduce
an extractor for mr->ibmr.page_size. Normal usage in the kernel
has page_mask masking out offset in page rather than masking out
the page number. The rxe driver had reversed that which was confusing.
Implicitly there can be a per mr page_size which was not uniformly
supported.
Link: https://lore.kernel.org/r/20230119235936.19728-6-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Isolate mr specific code from atomic_write_reply() in rxe_resp.c into
a subroutine rxe_mr_do_atomic_write() in rxe_mr.c.
Check length for atomic write operation.
Make iova_to_vaddr() static.
Link: https://lore.kernel.org/r/20230119235936.19728-5-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Isolate mr specific code from atomic_reply() in rxe_resp.c into
a subroutine rxe_mr_do_atomic_op() in rxe_mr.c.
Minor cleanups to rxe_check_range() and iova_to_vaddr().
Move enum resp_state to rxe.h
Link: https://lore.kernel.org/r/20230119235936.19728-4-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
As suggested by Cong, introduce a tracepoint for all ->sk_data_ready()
callback implementations. For example:
<...>
iperf-609 [002] ..... 70.660425: sk_data_ready: family=2 protocol=6 func=sock_def_readable
iperf-609 [002] ..... 70.660436: sk_data_ready: family=2 protocol=6 func=sock_def_readable
<...>
Suggested-by: Cong Wang <cong.wang@bytedance.com>
Signed-off-by: Peilin Ye <peilin.ye@bytedance.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
If you create MRs more than 0x10000 times after loading the module,
responder starts to reply NAKs for RDMA/Atomic operations because of rkey
violation detected in check_rkey(). The root cause is that rkeys are
incremented each time a new MR is created and the value overflows into the
range reserved for MWs.
This commit also increases the value of RXE_MAX_MW that has been limited
unlike other parameters.
Fixes: 0994a1bcd5 ("RDMA/rxe: Bump up default maximum values used via uverbs")
Link: https://lore.kernel.org/r/20221220080848.253785-2-matsuda-daisuke@fujitsu.com
Signed-off-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Tested-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
ibv_query_device() has reported incorrect device attributes, which are
actually not used by the device. Make the constants correspond with the
attributes shown to users.
Fixes: 3ccffe8abf ("RDMA/rxe: Move max_elem into rxe_type_info")
Fixes: 3225717f6d ("RDMA/rxe: Replace red-black trees by xarrays")
Link: https://lore.kernel.org/r/20221220080848.253785-1-matsuda-daisuke@fujitsu.com
Signed-off-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Fix two build warnings on 32 bit platforms
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQRRRCHOFoQz/8F5bUaFwuHvBreFYQUCY50UewAKCRCFwuHvBreF
YazuAQCHY/yOiziPBqNvI9jSb/MVGh1oaZQZRTkt5fOvLSGt8gD/Rv30/h6EcKfG
S5A6eQ3qrVzEhp7P8mST2Q/MkGmQ+AQ=
=Qw7m
-----END PGP SIGNATURE-----
Merge tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma
Pull rdma fixes from Jason Gunthorpe:
"Fix two build warnings on 32 bit platforms
It seems the linux-next CI and 0-day bot are not testing enough 32 bit
configurations, as soon as you merged the rdma pull request there were
two instant reports of warnings on these sytems that I would have
thought should have been covered by time in linux-next
Anyhow, here are the fixes so people don't hit problems with -Werror"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma:
RDMA/siw: Fix pointer cast warning
RDMA/rxe: Fix compile warnings on 32-bit
The previous build fix left a remaining issue in configurations with
64-bit dma_addr_t on 32-bit architectures:
drivers/infiniband/sw/siw/siw_qp_tx.c: In function 'siw_get_pblpage':
drivers/infiniband/sw/siw/siw_qp_tx.c:32:37: error: cast to pointer from integer of different size [-Werror=int-to-pointer-cast]
32 | return virt_to_page((void *)paddr);
| ^
Use the same double cast here that the driver uses elsewhere to convert
between dma_addr_t and void*.
Fixes: 0d1b756acf ("RDMA/siw: Pass a pointer to virt_to_page()")
Link: https://lore.kernel.org/r/20221215170347.2612403-1-arnd@kernel.org
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Bernard Metzler <bmt@zurich.ibm.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Usual size of updates, a new driver a most of the bulk focusing on rxe:
- Usual typos, style, and language updates
- Driver updates for mlx5, irdma, siw, rts, srp, hfi1, hns, erdma, mlx4, srp
- Lots of RXE updates
* Improve reply error handling for bad MR operations
* Code tidying
* Debug printing uses common loggers
* Remove half implemented RD related stuff
* Support IBA's recently defined Atomic Write and Flush operations
- erdma support for atomic operations
- New driver "mana" for Ethernet HW available in Azure VMs. This driver
only supports DPDK
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQRRRCHOFoQz/8F5bUaFwuHvBreFYQUCY5eIggAKCRCFwuHvBreF
YeX7AP9+l5Y9J48OmK7y/YgADNo9g05agXp3E8EuUDmBU+PREgEAigdWaJVf2oea
IctVja0ApLW5W+wsFt8Qh+V4PMiYTAM=
=Q5V+
-----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:
"Usual size of updates, a new driver, and most of the bulk focusing on
rxe:
- Usual typos, style, and language updates
- Driver updates for mlx5, irdma, siw, rts, srp, hfi1, hns, erdma,
mlx4, srp
- Lots of RXE updates:
* Improve reply error handling for bad MR operations
* Code tidying
* Debug printing uses common loggers
* Remove half implemented RD related stuff
* Support IBA's recently defined Atomic Write and Flush operations
- erdma support for atomic operations
- New driver 'mana' for Ethernet HW available in Azure VMs. This
driver only supports DPDK"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (122 commits)
IB/IPoIB: Fix queue count inconsistency for PKEY child interfaces
RDMA: Add missed netdev_put() for the netdevice_tracker
RDMA/rxe: Enable RDMA FLUSH capability for rxe device
RDMA/cm: Make QP FLUSHABLE for supported device
RDMA/rxe: Implement flush completion
RDMA/rxe: Implement flush execution in responder side
RDMA/rxe: Implement RC RDMA FLUSH service in requester side
RDMA/rxe: Extend rxe packet format to support flush
RDMA/rxe: Allow registering persistent flag for pmem MR only
RDMA/rxe: Extend rxe user ABI to support flush
RDMA: Extend RDMA kernel verbs ABI to support flush
RDMA: Extend RDMA user ABI to support flush
RDMA/rxe: Fix incorrect responder length checking
RDMA/rxe: Fix oops with zero length reads
RDMA/mlx5: Remove not-used IB_FLOW_SPEC_IB define
RDMA/hns: Fix XRC caps on HIP08
RDMA/hns: Fix error code of CMD
RDMA/hns: Fix page size cap from firmware
RDMA/hns: Fix PBL page MTR find
RDMA/hns: Fix AH attr queried by query_qp
...
Now we are ready to enable RDMA FLUSH capability for RXE.
It can support Global Visibility and Persistence placement types.
Link: https://lore.kernel.org/r/20221206130201.30986-11-lizhijian@fujitsu.com
Reviewed-by: Zhu Yanjun <zyjzyj2000@gmail.com>
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Per IBA SPEC, FLUSH will ack in rdma read response with 0 length.
Use IB_WC_FLUSH (aka IB_UVERBS_WC_FLUSH) code to tell userspace a FLUSH
completion.
Link: https://lore.kernel.org/r/20221206130201.30986-9-lizhijian@fujitsu.com
Reviewed-by: Zhu Yanjun <zyjzyj2000@gmail.com>
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Only the requested placement types that also registered in the destination
memory region are acceptable.
Otherwise, responder will also reply NAK "Remote Access Error" if it
found a placement type violation.
We will persist data via arch_wb_cache_pmem(), which could be
architecture specific.
This commit also adds 2 helpers to update qp.resp from the incoming packet.
Link: https://lore.kernel.org/r/20221206130201.30986-8-lizhijian@fujitsu.com
Reviewed-by: Zhu Yanjun <zyjzyj2000@gmail.com>
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Extend rxe opcode tables, headers, helper and constants to support
flush operations.
Refer to the IBA A19.4.1 for more FETH definition details
Link: https://lore.kernel.org/r/20221206130201.30986-6-lizhijian@fujitsu.com
Reviewed-by: Zhu Yanjun <zyjzyj2000@gmail.com>
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Memory region could support at most 2 flush access flags:
IB_ACCESS_FLUSH_PERSISTENT and IB_ACCESS_FLUSH_GLOBAL
But we only allow user to register persistent flush flags to the pmem MR
where it has the ability of persisting data across power cycles.
So registering a persistent access flag to a non-pmem MR will be rejected.
Link: https://lore.kernel.org/r/20221206130201.30986-5-lizhijian@fujitsu.com
CC: Dan Williams <dan.j.williams@intel.com>
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The code in rxe_resp.c at check_length() is incorrect as it compares
pkt->opcode an 8 bit value against various mask bits which are all higher
than 256 so nothing is ever reported.
This patch rewrites this to compare against pkt->mask which is
correct. However this now exposes another error. For UD send packets the
value of the pmtu cannot be determined from qp->mtu. All that is required
here is to later check if the payload fits into the posted receive buffer
in that case.
Fixes: 837a55847e ("RDMA/rxe: Implement packet length validation on responder")
Link: https://lore.kernel.org/r/20221208210945.28607-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Reviewed-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
-----BEGIN PGP SIGNATURE-----
iQFSBAABCAA8FiEEq68RxlopcLEwq+PEeb4+QwBBGIYFAmONI6weHHRvcnZhbGRz
QGxpbnV4LWZvdW5kYXRpb24ub3JnAAoJEHm+PkMAQRiG9xgH/jqXGuMoO1ikfmGb
7oY0W/f69G9V/e0DxFLvnIjhFgCUzdnNsmD4jQJA4x6QsxwLWuvpI282Ez+bHV5T
U4RPsxJZIIMsXE2lKM9BRgeLzDdCt0aK4Pj+3x2x7NZC5cWFSQ8PyQJkCwg+0PQo
u8Ly+GO8c4RUMf4/rrAZQq16qZUqGDaGm1EJhtSoa+KiR81LmUUmbDIK9Mr53rmQ
wou+95XhibwMWr17WgXA28bTgYqn9UGr67V3qvTH2LC7GW8BCoKvn+3wh6TVhlWj
dsWplXgcOP0/OHvSC5Sb1Uibk5Gx3DlIzYa6OfNZQuZ5xmQqm9kXjW8lmYpWFHy/
38/5HWc=
=EuoA
-----END PGP SIGNATURE-----
Merge tag 'v6.1-rc8' into rdma.git for-next
For dependencies in following patches
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
GUP now supports reliable R/O long-term pinning in COW mappings, such
that we break COW early. MAP_SHARED VMAs only use the shared zeropage so
far in one corner case (DAXFS file with holes), which can be ignored
because GUP does not support long-term pinning in fsdax (see
check_vma_flags()).
Consequently, FOLL_FORCE | FOLL_WRITE | FOLL_LONGTERM is no longer required
for reliable R/O long-term pinning: FOLL_LONGTERM is sufficient. So stop
using FOLL_FORCE, which is really only for ptrace access.
Link: https://lkml.kernel.org/r/20221116102659.70287-13-david@redhat.com
Signed-off-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Jason Gunthorpe <jgg@nvidia.com>
Cc: Bernard Metzler <bmt@zurich.ibm.com>
Cc: Leon Romanovsky <leon@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
There is a null-ptr-deref when mount.cifs over rdma:
BUG: KASAN: null-ptr-deref in rxe_qp_do_cleanup+0x2f3/0x360 [rdma_rxe]
Read of size 8 at addr 0000000000000018 by task mount.cifs/3046
CPU: 2 PID: 3046 Comm: mount.cifs Not tainted 6.1.0-rc5+ #62
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.14.0-1.fc3
Call Trace:
<TASK>
dump_stack_lvl+0x34/0x44
kasan_report+0xad/0x130
rxe_qp_do_cleanup+0x2f3/0x360 [rdma_rxe]
execute_in_process_context+0x25/0x90
__rxe_cleanup+0x101/0x1d0 [rdma_rxe]
rxe_create_qp+0x16a/0x180 [rdma_rxe]
create_qp.part.0+0x27d/0x340
ib_create_qp_kernel+0x73/0x160
rdma_create_qp+0x100/0x230
_smbd_get_connection+0x752/0x20f0
smbd_get_connection+0x21/0x40
cifs_get_tcp_session+0x8ef/0xda0
mount_get_conns+0x60/0x750
cifs_mount+0x103/0xd00
cifs_smb3_do_mount+0x1dd/0xcb0
smb3_get_tree+0x1d5/0x300
vfs_get_tree+0x41/0xf0
path_mount+0x9b3/0xdd0
__x64_sys_mount+0x190/0x1d0
do_syscall_64+0x35/0x80
entry_SYSCALL_64_after_hwframe+0x46/0xb0
The root cause of the issue is the socket create failed in
rxe_qp_init_req().
So move the reset rxe_qp_do_cleanup() after the NULL ptr check.
Fixes: 8700e3e7c4 ("Soft RoCE driver")
Link: https://lore.kernel.org/r/20221122151437.1057671-1-zhangxiaoxu5@huawei.com
Signed-off-by: Zhang Xiaoxu <zhangxiaoxu5@huawei.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Correct the mistake, mr is obviously NULL in this code path.
Fixes: 2778b72b1d ("RDMA/rxe: Replace pr_xxx by rxe_dbg_xxx in rxe_mr.c")
Link: https://lore.kernel.org/r/Y3eeJW0AdyJYhYyQ@kili
Reported-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
rxe_mr_cleanup() which tries to free mr->map again will be called when
rxe_mr_init_user() fails:
CPU: 0 PID: 4917 Comm: rdma_flush_serv Kdump: loaded Not tainted 6.1.0-rc1-roce-flush+ #25
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.0-0-gd239552ce722-prebuilt.qemu.org 04/01/2014
Call Trace:
<TASK>
dump_stack_lvl+0x45/0x5d
panic+0x19e/0x349
end_report.part.0+0x54/0x7c
kasan_report.cold+0xa/0xf
rxe_mr_cleanup+0x9d/0xf0 [rdma_rxe]
__rxe_cleanup+0x10a/0x1e0 [rdma_rxe]
rxe_reg_user_mr+0xb7/0xd0 [rdma_rxe]
ib_uverbs_reg_mr+0x26a/0x480 [ib_uverbs]
ib_uverbs_handler_UVERBS_METHOD_INVOKE_WRITE+0x1a2/0x250 [ib_uverbs]
ib_uverbs_cmd_verbs+0x1397/0x15a0 [ib_uverbs]
This issue was firstly exposed since commit b18c7da63f ("RDMA/rxe: Fix
memory leak in error path code") and then we fixed it in commit
8ff5f5d9d8 ("RDMA/rxe: Prevent double freeing rxe_map_set()") but this
fix was reverted together at last by commit 1e75550648 (Revert
"RDMA/rxe: Create duplicate mapping tables for FMRs")
Simply let rxe_mr_cleanup() always handle freeing the mr->map once it is
successfully allocated.
Fixes: 1e75550648 ("Revert "RDMA/rxe: Create duplicate mapping tables for FMRs"")
Link: https://lore.kernel.org/r/1667099073-2-1-git-send-email-lizhijian@fujitsu.com
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The rdma_rxe driver does not actually support the reliable datagram
transport but contains a variable with RD opcodes in driver code. And
this variable is never used. So remove it.
Link: https://lore.kernel.org/r/20221112023537.432912-1-yanjun.zhu@intel.com
Signed-off-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
A malicious user may write undefined values into memory mapped completion
queue elements status or opcode. Undefined status or opcode values will
result in out-of-bounds access to an array mapping siw internal
representation of opcode and status to RDMA core representation when
reaping CQ elements. While siw detects those undefined values, it did not
correctly set completion status to a defined value, thus defeating the
whole purpose of the check.
This bug leads to the following Smatch static checker warning:
drivers/infiniband/sw/siw/siw_cq.c:96 siw_reap_cqe()
error: buffer overflow 'map_cqe_status' 10 <= 21
Fixes: bdf1da5df9 ("RDMA/siw: Fix immediate work request flush to completion queue")
Link: https://lore.kernel.org/r/20221115170747.1263298-1-bmt@zurich.ibm.com
Reported-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Bernard Metzler <bmt@zurich.ibm.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Replace calls to pr_xxx() in rxe.c with rxe_dbg_xxx().
Calls with a rxe device not yet in scope are left as is.
Link: https://lore.kernel.org/r/20221103171013.20659-15-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Replace (some) calls to pr_xxx() in rxe_net.c with rxe_dbg_xxx().
Calls with a rxe device not yet in scope are left as is.
Link: https://lore.kernel.org/r/20221103171013.20659-7-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The function check_length() is supposed to check the length of inbound
packets on responder, but it actually has been a stub since the driver was
born. Let it check the payload length and the DMA length.
Signed-off-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Link: https://lore.kernel.org/r/20221107055338.357184-1-matsuda-daisuke@fujitsu.com
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Acked-by: Zhu Yanjun <zyjzyj2000@gmail.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Correctly set send queue element opcode during immediate work request
flushing in post sendqueue operation, if the QP is in ERROR state.
An undefined ocode value results in out-of-bounds access to an array
for mapping the opcode between siw internal and RDMA core representation
in work completion generation. It resulted in a KASAN BUG report
of type 'global-out-of-bounds' during NFSoRDMA testing.
This patch further fixes a potential case of a malicious user which may
write undefined values for completion queue elements status or opcode,
if the CQ is memory mapped to user land. It avoids the same out-of-bounds
access to arrays for status and opcode mapping as described above.
Fixes: 303ae1cdfd ("rdma/siw: application interface")
Fixes: b0fff7317b ("rdma/siw: completion queue methods")
Reported-by: Olga Kornievskaia <kolga@netapp.com>
Reviewed-by: Tom Talpey <tom@talpey.com>
Signed-off-by: Bernard Metzler <bmt@zurich.ibm.com>
Link: https://lore.kernel.org/r/20221107145057.895747-1-bmt@zurich.ibm.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Instead of 'goto and return', just return directly to
simplify the error handling, and avoid some unnecessary
return value check.
Link: https://lore.kernel.org/r/20221028075053.3990467-1-xuhaoyue1@hisilicon.com
Signed-off-by: Yunsheng Lin <linyunsheng@huawei.com>
Signed-off-by: Haoyue Xu <xuhaoyue1@hisilicon.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
mr->map_shift is set to ilog2(RXE_BUF_PER_MAP) in both rxe_mr_init() and
rxe_mr_alloc() so remove the duplicate one in rxe_mr_init().
Link: https://lore.kernel.org/r/1666855893-145-1-git-send-email-yangx.jy@fujitsu.com
Signed-off-by: Xiao Yang <yangx.jy@fujitsu.com>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
We should reject the requests with access flags that is not registered by
MR/MW. For example, lookup_mr() should return NULL when requested access
is 0x03 and mr->access is 0x01.
Link: https://lore.kernel.org/r/20220927055337.22630-2-lizhijian@fujitsu.com
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The subroutine rxe_do_task() is only called in rxe_task.c. This patch
makes it static and renames it do_task().
Link: https://lore.kernel.org/r/20221021200118.2163-6-rpearsonhpe@gmail.com
Signed-off-by: Ian Ziemba <ian.ziemba@hpe.com>
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Split rxe_run_task(task, sched) into rxe_run_task(task) and
rxe_sched_task(task).
Link: https://lore.kernel.org/r/20221021200118.2163-5-rpearsonhpe@gmail.com
Signed-off-by: Ian Ziemba <ian.ziemba@hpe.com>
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The name field in struct rxe_task is never used. This patch removes it.
Link: https://lore.kernel.org/r/20221021200118.2163-4-rpearsonhpe@gmail.com
Signed-off-by: Ian Ziemba <ian.ziemba@hpe.com>
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The calls to spin_lock_init() for the tasklet spinlocks in
rxe_qp_init_misc() are redundant since they are intiialized in
rxe_init_task(). This patch removes them.
Link: https://lore.kernel.org/r/20221021200118.2163-3-rpearsonhpe@gmail.com
Signed-off-by: Ian Ziemba <ian.ziemba@hpe.com>
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Before the testing, we already passed it to rxe_mr_copy() where mr could
be dereferenced. so this checking is not needed.
The only way that mr is NULL is when it reaches below line 780 with
'qp->resp.mr = NULL', which is not possible in Bob's explanation[1].
778 if (res->state == rdatm_res_state_new) {
779 if (!res->replay) {
780 mr = qp->resp.mr;
781 qp->resp.mr = NULL;
782 } else {
[1] https://lore.kernel.org/lkml/30ff25c4-ce66-eac4-eaa2-64c0db203a19@gmail.com/
Link: https://lore.kernel.org/r/1666582315-2-1-git-send-email-lizhijian@fujitsu.com
CC: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Requesting nodes do not handle a reported error correctly if it is
generated in the middle of multi-packet Read responses, and the node tries
to resend the request endlessly. Let completer terminate the connection in
that case.
Link: https://lore.kernel.org/r/20221013014724.3786212-2-matsuda-daisuke@fujitsu.com
Signed-off-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Currently, responder can reply packets with invalid payloads if it fails
to copy messages to the packets. Add an error handling in read_reply() to
inform a requesting node of the failure.
Link: https://lore.kernel.org/r/20221013014724.3786212-1-matsuda-daisuke@fujitsu.com
Suggested-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
The member 'type' is included in both struct rxe_mr and struct ib_mr
so remove the duplicate one of struct rxe_mr.
Signed-off-by: Xiao Yang <yangx.jy@fujitsu.com>
Link: https://lore.kernel.org/r/20221021134513.17730-1-yangx.jy@fujitsu.com
Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Incoming packets to rxe are passed from UDP layer using an encapsulation
socket. If there are any clients reachable to a node, they can invoke the
encapsulation handler arbitrarily by sending malicious or irrelevant
packets. This can potentially cause a message overflow and a subsequent
slowdown on the node.
Signed-off-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Link: https://lore.kernel.org/r/20220929080023.304242-1-matsuda-daisuke@fujitsu.com
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
In include/uapi/rdma/rdma_user_rxe.h there are redundant copies of num_sge
in the rxe_send_wr, rxe_recv_wqe, and rxe_dma_info. Only the ones in
rxe_dma_info are actually used by the rxe kernel driver.
The userspace would set these values, but the kernel never read them.
This change has no affect on the current ABI and new or old versions of
rdma-core operate correctly with new or old versions of the kernel rxe
driver.
Link: https://lore.kernel.org/r/20220913222716.18335-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Currently in resize_finish() in rxe_queue.c there is a loop which copies
the entries in the original queue into a newly allocated queue. The
termination logic for this loop is incorrect. The call to
queue_next_index() updates cons but has no effect on whether the queue is
empty. So if the queue starts out empty nothing is copied but if it is not
then the loop will run forever. This patch changes the loop to compare the
value of cons to the original producer index.
Fixes: ae6e843fe0 ("RDMA/rxe: Add memory barriers to kernel queues")
Link: https://lore.kernel.org/r/20220825221446.6512-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Move setting of pd in mr objects ahead of any possible errors so that it
will always be set in rxe_mr_cleanup() to avoid seg faults when
rxe_put(mr_pd(mr)) is called.
Fixes: cf40367961 ("RDMA/rxe: Move mr cleanup code to rxe_mr_cleanup()")
Link: https://lore.kernel.org/r/20220805183153.32007-2-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Most code in send_ack() and send_atomic_ack() are duplicate, move them to
a new helper send_common_ack().
In newer IBA spec, some opcodes require acknowledge with a zero-length
read response, with this new helper, we can easily implement it later.
Link: https://lore.kernel.org/r/1659335010-2-1-git-send-email-lizhijian@fujitsu.com
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
rxe_mr and ib_mr have interchangeable members. Remove device specific
members and use ones in the generic struct. Both 'iova' and 'length' are
filled in ib_uverbs or ib_core layer after MR registration.
Signed-off-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Link: https://lore.kernel.org/r/20220921080844.1616883-2-matsuda-daisuke@fujitsu.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Delay QP destroy completion until all siw references to QP are
dropped. The calling RDMA core will free QP structure after
successful return from siw_qp_destroy() call, so siw must not
hold any remaining reference to the QP upon return.
A use-after-free was encountered in xfstest generic/460, while
testing NFSoRDMA. Here, after a TCP connection drop by peer,
the triggered siw_cm_work_handler got delayed until after
QP destroy call, referencing a QP which has already freed.
Fixes: 303ae1cdfd ("rdma/siw: application interface")
Reported-by: Olga Kornievskaia <kolga@netapp.com>
Signed-off-by: Bernard Metzler <bmt@zurich.ibm.com>
Link: https://lore.kernel.org/r/20220920082503.224189-1-bmt@zurich.ibm.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
For header and trailer/padding processing, siw did not consume new
skb data until minimum amount present to fill current header or trailer
structure, including potential payload padding. Not consuming any
data during upcall may cause a receive stall, since tcp_read_sock()
is not upcalling again if no new data arrive.
A NFSoRDMA client got stuck at RDMA Write reception of unaligned
payload, if the current skb did contain only the expected 3 padding
bytes, but not the 4 bytes CRC trailer. Expecting 4 more bytes already
arrived in another skb, and not consuming those 3 bytes in the current
upcall left the Write incomplete, waiting for the CRC forever.
Fixes: 8b6a361b8c ("rdma/siw: receive path")
Reported-by: Olga Kornievskaia <kolga@netapp.com>
Tested-by: Olga Kornievskaia <kolga@netapp.com>
Signed-off-by: Bernard Metzler <bmt@zurich.ibm.com>
Link: https://lore.kernel.org/r/20220920081202.223629-1-bmt@zurich.ibm.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Functions that work on a pointer to virtual memory such as
virt_to_pfn() and users of that function such as
virt_to_page() are supposed to pass a pointer to virtual
memory, ideally a (void *) or other pointer. However since
many architectures implement virt_to_pfn() as a macro,
this function becomes polymorphic and accepts both a
(unsigned long) and a (void *).
If we instead implement a proper virt_to_pfn(void *addr)
function the following happens (occurred on arch/arm):
drivers/infiniband/sw/siw/siw_qp_tx.c:32:23: warning: incompatible
integer to pointer conversion passing 'dma_addr_t' (aka 'unsigned int')
to parameter of type 'const void *' [-Wint-conversion]
drivers/infiniband/sw/siw/siw_qp_tx.c:32:37: warning: passing argument
1 of 'virt_to_pfn' makes pointer from integer without a cast
[-Wint-conversion]
drivers/infiniband/sw/siw/siw_qp_tx.c:538:36: warning: incompatible
integer to pointer conversion passing 'unsigned long long'
to parameter of type 'const void *' [-Wint-conversion]
Fix this with an explicit cast. In one case where the SIW
SGE uses an unaligned u64 we need a double cast modifying the
virtual address (va) to a platform-specific uintptr_t before
casting to a (void *).
Fixes: b9be6f18cf ("rdma/siw: transmit path")
Cc: linux-rdma@vger.kernel.org
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://lore.kernel.org/r/20220902215918.603761-1-linus.walleij@linaro.org
Signed-off-by: Leon Romanovsky <leon@kernel.org>
The SoftiWARP Kconfig is missing "select" for CRYPTO and CRYPTO_CRC32C.
In addition, it improperly "depends on" LIBCRC32C, this should be a
"select", similar to net/sctp and others. As a dependency, SIW fails
to appear in generic configurations.
Link: https://lore.kernel.org/r/d366bf02-3271-754f-fc68-1a84016d0e19@talpey.com
Signed-off-by: Tom Talpey <tom@talpey.com>
Acked-by: Bernard Metzler <bmt@zurich.ibm.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
An incoming Read request causes multiple Read responses. If a user MR to
copy data from is unavailable or responder cannot send a reply, then the
error messages can be printed for each response attempt, resulting in
message overflow.
Link: https://lore.kernel.org/r/20220829071218.1639065-1-matsuda-daisuke@fujitsu.com
Signed-off-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
The member variable obj in struct rxe_task is not needed.
So remove it to save memory.
Link: https://lore.kernel.org/r/20220822011615.805603-4-yanjun.zhu@linux.dev
Signed-off-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
When sock_create_kern in the function rxe_qp_init_req fails,
qp->sk is set to NULL.
Then the function rxe_create_qp will call rxe_qp_do_cleanup
to handle allocated resource.
Before handling qp->sk, this variable should be checked.
Fixes: 8700e3e7c4 ("Soft RoCE driver")
Link: https://lore.kernel.org/r/20220822011615.805603-3-yanjun.zhu@linux.dev
Signed-off-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
When rxe_queue_init in the function rxe_qp_init_req fails,
both qp->req.task.func and qp->req.task.arg are not initialized.
Because of creation of qp fails, the function rxe_create_qp will
call rxe_qp_do_cleanup to handle allocated resource.
Before calling __rxe_do_task, both qp->req.task.func and
qp->req.task.arg should be checked.
Fixes: 8700e3e7c4 ("Soft RoCE driver")
Link: https://lore.kernel.org/r/20220822011615.805603-2-yanjun.zhu@linux.dev
Reported-by: syzbot+ab99dc4c6e961eed8b8e@syzkaller.appspotmail.com
Signed-off-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Commit 1e75550648 ("Revert "RDMA/rxe: Create duplicate mapping tables for
FMRs"") brought back the member 'va' to struct rxe_mr. However, it is
actually used by nobody and thus can be removed.
Fixes: 1e75550648 ("Revert "RDMA/rxe: Create duplicate mapping tables for FMRs"")
Link: https://lore.kernel.org/r/20220829012335.1212697-1-matsuda-daisuke@fujitsu.com
Signed-off-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
In the function rxe_create_qp(), rxe_qp_from_init() is called to
initialize qp, internally things like the spin locks are not setup until
rxe_qp_init_req().
If an error occures before this point then the unwind will call
rxe_cleanup() and eventually to rxe_qp_do_cleanup()/rxe_cleanup_task()
which will oops when trying to access the uninitialized spinlock.
Move the spinlock initializations earlier before any failures.
Fixes: 8700e3e7c4 ("Soft RoCE driver")
Link: https://lore.kernel.org/r/20220731063621.298405-1-yanjun.zhu@linux.dev
Reported-by: syzbot+833061116fa28df97f3b@syzkaller.appspotmail.com
Signed-off-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Currently the requester can continue to process send wqes after an local
qp operation error is detected because the setting of the qp state to the
error state is deferred until later. This patch splits the qp state for
the completer and requester into two separate states and sets
qp->req.state = QP_STATE_ERROR as soon as the error is detected before
another wqe can be executed.
Link: https://lore.kernel.org/r/1658307368-1851-4-git-send-email-lizhijian@fujitsu.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
As per IBTA specification, all subsequent WQEs while QP is in error state
should be completed with a flush error.
Here we check QP_STATE_ERROR after req_next_wqe() so that rxe_completer()
has chance to be called where it will set CQ state to FLUSH ERROR and the
completion can associate with its WQE.
Link: https://lore.kernel.org/r/1658307368-1851-3-git-send-email-lizhijian@fujitsu.com
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Previously, if user space keeps sending abnormal wqe, queue.index will
keep increasing while qp->req.wqe_index doesn't. Once
qp->req.wqe_index==queue.index in next round, req_next_wqe() will treat
queue as empty. In such case, no new completion would be generated.
Update wqe_index for each wqe completion so that req_next_wqe() can get
next wqe properly.
Link: https://lore.kernel.org/r/1658307368-1851-2-git-send-email-lizhijian@fujitsu.com
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
In rxe_req.c replace calls to __rxe_do_task() by calls to rxe_run_task(..,
0). Using __rxe_do_task is an error because the completer tasklet is not
designed to be re-entrant and __rxe_do_task() should only be called when
it is clear that no one else could be calling the completer tasklet as is
the case in rxe_qp.c where this call is used in safe environments.
Link: https://lore.kernel.org/r/20220630190425.2251-10-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Limit the maximum number of calls to each tasklet from rxe_do_task()
before yielding the cpu. When the limit is reached reschedule the tasklet
and exit the calling loop. This patch prevents one tasklet from consuming
100% of a cpu core and causing a deadlock or soft lockup.
Link: https://lore.kernel.org/r/20220630190425.2251-9-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Make changes to the three tasklets so that the exit logic from each is the
same. This makes the code easier to understand.
Link: https://lore.kernel.org/r/20220630190425.2251-8-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Currently the completer tasklet when retransmit timer or the rnr timer
fires the same flag (qp->req.need_retry) is set so that if either timer
fires it will attempt to perform a retry flow on the send queue. This has
the effect of responding to an RNR NAK at the first retransmit timer event
which might not allow the requested rnr timeout.
This patch adds a new flag (qp->req.wait_for_rnr_timer) which, if set,
prevents a retry flow until the rnr nak timer fires.
This patch fixes rnr retry errors which can be observed by running the
pyverbs test_rdmacm_async_traffic_external_qp multiple times. With this
patch applied they do not occur.
Link: https://lore.kernel.org/linux-rdma/a8287823-1408-4273-bc22-99a0678db640@gmail.com/
Link: https://lore.kernel.org/linux-rdma/2bafda9e-2bb6-186d-12a1-179e8f6a2678@talpey.com/
Fixes: 8700e3e7c4 ("Soft RoCE driver")
Link: https://lore.kernel.org/r/20220630190425.2251-6-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The code thc that decides whether to defer execution of a wqe in
rxe_requester.c is isolated into a subroutine rxe_is_fenced() and removed
from the call to req_next_wqe(). The condition whether a wqe should be
fenced is changed to comply with the IBA. Currently an operation is fenced
if the fence bit is set in the wqe flags and the last wqe has not
completed. For normal operations the IBA actually only requires that the
last read or atomic operation is complete.
Link: https://lore.kernel.org/r/20220630190425.2251-2-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Reviewed-by: Li Zhijian <lizhijian@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The 'rkey' input can be an lkey or rkey, and in rxe the lkey or rkey have
the same value, including the variant bits.
So, if mr->rkey is set, compare the invalidate key with it, otherwise
compare with the mr->lkey.
Since we already did a lookup on the non-varient bits to get this far, the
check's only purpose is to confirm that the wqe has the correct variant
bits.
Fixes: 001345339f ("RDMA/rxe: Separate HW and SW l/rkeys")
Link: https://lore.kernel.org/r/20220707073006.328737-1-haris.phnx@gmail.com
Signed-off-by: Md Haris Iqbal <haris.phnx@gmail.com>
Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
There is a spelling mistake (writeable) in function rxe_check_bind_mw.
Fix it.
Signed-off-by: Zhang Jiaming <jiaming@nfschina.com>
Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
The qp parameter in free_rd_atomic_resource() has become
unused so remove it directly.
Fixes: 15ae1375ea ("RDMA/rxe: Fix qp reference counting for atomic ops")
Link: https://lore.kernel.org/all/20220708035547.6592-1-yangx.jy@fujitsu.com/
Signed-off-by: Xiao Yang <yangx.jy@fujitsu.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
This parameter had been deprecated since below commit:
1a7085b342 ("RDMA/rxe: Skip adjusting remote addr for write in retry operation")
Link: https://lore.kernel.org/r/20220715035340.1900168-1-lizhijian@fujitsu.com
Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
It's redundant to prepare resources for Read and Atomic
requests by different functions. Replace them by a common
rxe_prepare_res() with different parameters. In addition,
the common rxe_prepare_res() can also be used by new Flush
and Atomic Write requests in the future.
Link: https://lore.kernel.org/r/20220705145212.12014-1-yangx.jy@fujitsu.com
Signed-off-by: Xiao Yang <yangx.jy@fujitsu.com>
Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
The function rxe_create_qp calls rxe_qp_from_init. If some error
occurs, the error handler of function rxe_qp_from_init will set
both scq and rcq to NULL.
Then rxe_create_qp calls rxe_put to handle qp. In the end,
rxe_qp_do_cleanup is called by rxe_put. rxe_qp_do_cleanup directly
accesses scq and rcq before checking them. This will cause
null-ptr-deref error.
The call graph is as below:
rxe_create_qp {
...
rxe_qp_from_init {
...
err1:
...
qp->rcq = NULL; <---rcq is set to NULL
qp->scq = NULL; <---scq is set to NULL
...
}
qp_init:
rxe_put{
...
rxe_qp_do_cleanup {
...
atomic_dec(&qp->scq->num_wq); <--- scq is accessed
...
atomic_dec(&qp->rcq->num_wq); <--- rcq is accessed
}
}
Fixes: 4703b4f0d9 ("RDMA/rxe: Enforce IBA C11-17")
Link: https://lore.kernel.org/r/20220705225414.315478-1-yanjun.zhu@linux.dev
Signed-off-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com>
Reviewed-by: Md Haris Iqbal <haris.iqbal@ionos.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
If siw_recv_mpa_rr returns -EAGAIN, it means that the MPA reply hasn't
been received completely, and should not report IW_CM_EVENT_CONNECT_REPLY
in this case. This may trigger a call trace in iw_cm. A simple way to
trigger this:
server: ib_send_lat
client: ib_send_lat -R <server_ip>
The call trace looks like this:
kernel BUG at drivers/infiniband/core/iwcm.c:894!
invalid opcode: 0000 [#1] PREEMPT SMP NOPTI
<...>
Workqueue: iw_cm_wq cm_work_handler [iw_cm]
Call Trace:
<TASK>
cm_work_handler+0x1dd/0x370 [iw_cm]
process_one_work+0x1e2/0x3b0
worker_thread+0x49/0x2e0
? rescuer_thread+0x370/0x370
kthread+0xe5/0x110
? kthread_complete_and_exit+0x20/0x20
ret_from_fork+0x1f/0x30
</TASK>
Fixes: 6c52fdc244 ("rdma/siw: connection management")
Link: https://lore.kernel.org/r/dae34b5fd5c2ea2bd9744812c1d2653a34a94c67.1657706960.git.chengyou@linux.alibaba.com
Signed-off-by: Cheng Xu <chengyou@linux.alibaba.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Comparison of 'cq' with NULL is useless since
'cq' is a result of container_of and cannot be NULL
in any reasonable scenario.
Found by Linux Verification Center (linuxtesting.org) with SVACE.
Fixes: 303ae1cdfd ("rdma/siw: application interface")
Link: https://lore.kernel.org/r/20220711151251.17089-1-strochuk@ispras.ru
Signed-off-by: Andrey Strachuk <strochuk@ispras.ru>
Acked-by: Bernard Metzler <bmt@zurich.ibm.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
rxe_verbs.h includes the file <rdma/rdma_user_rxe.h>. It should have been
<uapi/rdma/rdma_user_rxe.h>, however, it is not used and not required in
this file.
This patch removes the include statement.
Link: https://lore.kernel.org/r/20220630190425.2251-4-rpearsonhpe@gmail.com
Reported-by: Frank Zago <frank.zago@hpe.com>
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
rxe_queue.h currently includes <uapi/rdma/rdma_user_rxe.h> for a
definition of struct rxe_queue_buf. But it is only used as a pointer so
the definition is not needed.
This patch replaces the include statement with the declaration
struct rxe_queue_buf;
Link: https://lore.kernel.org/r/20220630190425.2251-5-rpearsonhpe@gmail.com
Reported-by: Frank Zago <frank.zago@hpe.com>
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The pyverbs test suite generates a few dmesg traces from intentional error
tests. This patch replaces those messages with pr_debug() calls which
improves the usefullness of the tests.
Link: https://lore.kernel.org/r/20220630190425.2251-3-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
When a local operation (invalidate mr, reg mr, bind mw) is finished there
will be no ack packet coming from a responder to cause the wqe to be
completed. This may happen anyway if a subsequent wqe performs
IO. Currently if the wqe is signalled the completer tasklet is scheduled
immediately but not otherwise.
This leads to a deadlock if the next wqe has the fence bit set in send
flags and the operation is not signalled. This patch removes the condition
that the wqe must be signalled in order to schedule the completer tasklet
which is the simplest fix for this deadlock and is fairly low cost. This
is the analog for local operations of always setting the ackreq bit in all
last or only request packets even if the operation is not signalled.
Link: https://lore.kernel.org/r/20220523223251.15350-1-rpearsonhpe@gmail.com
Reported-by: Jenny Hack <jhack@hpe.com>
Fixes: c1a411268a ("RDMA/rxe: Move local ops to subroutine")
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Make the execution of the atomic operation in rxe_atomic_reply()
conditional on res->replay and make duplicate_request() call into
rxe_atomic_reply() to merge the two flows. This is modeled on the behavior
of read reply. Delete the skb from the atomic responder resource since it
is no longer used. Adjust the reference counting of the qp in
send_atomic_ack() for this flow.
Fixes: 8700e3e7c4 ("Soft RoCE driver")
Link: https://lore.kernel.org/r/20220606143836.3323-6-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Move the saved original value to the atomic responder resource. This
replaces saving it in the qp. In preparation for merging the normal and
retry atomic responder flows.
Link: https://lore.kernel.org/r/20220606143836.3323-5-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Move the allocation of the atomic responder resource up into
rxe_atomic_reply() from send_atomic_ack(). In preparation for merging the
normal and retry atomic responder flows.
Link: https://lore.kernel.org/r/20220606143836.3323-4-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Add a responder state for atomic reply similar to read reply and rename
process_atomic() rxe_atomic_reply(). In preparation for merging the normal
and retry atomic responder flows.
Link: https://lore.kernel.org/r/20220606143836.3323-3-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Separate the code that prepares the atomic responder resource into a
subroutine. This is preparation for merging the normal and retry atomic
responder flows.
Link: https://lore.kernel.org/r/20220606143836.3323-2-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Currently the rdma_rxe driver has a security weakness due to giving
objects which are partially initialized indices allowing external actors
to gain access to them by sending packets which refer to their
index (e.g. qpn, rkey, etc) causing unpredictable results.
This patch adds a new API rxe_finalize(obj) which enables looking up pool
objects from indices using rxe_pool_get_index() for AH, QP, MR, and
MW. They are added in create verbs only after the objects are fully
initialized.
It also adds wait for completion to destroy/dealloc verbs to assure that
all references have been dropped before returning to rdma_core by
implementing a new rxe_pool API rxe_cleanup() which drops a reference to
the object and then waits for all other references to be dropped. When
the last reference is dropped the object is completed by kref. After that
it cleans up the object and if locally allocated frees the memory. In the
special case of address handle objects the delay is implemented separately
if the destroy_ah call is not sleepable.
Combined with deferring cleanup code to type specific cleanup routines
this allows all pending activity referring to objects to complete before
returning to rdma_core.
Link: https://lore.kernel.org/r/20220612223434.31462-2-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The pkt parameters in prepare_ack_packet(), send_ack() and
send_atomic_ack() have become useless by the following commits. So remove
them directly.
Fixes: bf139b58af ("RDMA/rxe: Remove unused pkt->offset")
Fixes: 3896bde92d ("RDMA/rxe: Fix extra copy in prepare_ack_packet")
Link: https://lore.kernel.org/r/20220623131627.18903-1-yangx.jy@fujitsu.com
Signed-off-by: Xiao Yang <yangx.jy@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Currently rxe_alloc checks ret to indicate error, but 1 is also a valid
return and just indicates that the allocation succeeded with a wrap.
Fix this by modifying the check to be < 0.
Link: https://lore.kernel.org/r/20220609070656.1446121-1-dzm91@hust.edu.cn
Fixes: 3225717f6d ("RDMA/rxe: Replace red-black trees by xarrays")
Signed-off-by: Dongliang Mu <mudongliangabcd@gmail.com>
Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Leon Romanovsky <leon@kernel.org>
-----BEGIN PGP SIGNATURE-----
iQFSBAABCAA8FiEEq68RxlopcLEwq+PEeb4+QwBBGIYFAmKKlIAeHHRvcnZhbGRz
QGxpbnV4LWZvdW5kYXRpb24ub3JnAAoJEHm+PkMAQRiGC3oH/iPm/fLG2sJut8My
sU0RC9K+6ESV5h2Qy6k00/lqKstlu4EvBjw4V8vYpx3Q2+hbSFMn2SeWqqqT3Lkk
Zb8KINCFuuyMtdCBb42PV0zhUf5pCQF7ocm/Ae4jllDHtPmqk3WJ6IGtZBK5JBlw
z6RR/wKt0y0MRj9eZyPyYjOee2L2vuVh4tgnexK/4L8g2ZtMMRThhvUzSMWG4zxR
STYYNp0uFcfT1Vt85+ODevFH4TvdECAj+SqAegN+seHLM17YY7M0/WiIYpxGRv8P
lIpDQl4PBU8EBkpI5hkpJ/3qPincbuVOMLsYfxFtpcjjG12vGjFp2krGpS3TedZQ
3mvaJ7c=
=vLke
-----END PGP SIGNATURE-----
Merge tag 'v5.18' into rdma.git for-next
Following patches have dependencies.
Resolve the merge conflict in
drivers/net/ethernet/mellanox/mlx5/core/main.c by keeping the new names
for the fs functions following linux-next:
https://lore.kernel.org/r/20220519113529.226bc3e2@canb.auug.org.au/
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Enable siw to attach to tunnel devices, there is no reason not to, siw
properly generates all packets already.
Link: https://lore.kernel.org/r/20220510143917.23735-1-bmt@zurich.ibm.com
Tested-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: Bernard Metzler <bmt@zurich.ibm.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Add a counter to keep track of the number of WQs connected to a CQ and
return an error if destroy_cq() is called while the counter is non zero.
Link: https://lore.kernel.org/r/20220421014042.26985-8-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Move code from rxe_dealloc_mw() to rxe_mw_cleanup() to allow flows which
hold a reference to mw to complete.
Link: https://lore.kernel.org/r/20220421014042.26985-7-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Move the code which tears down an mr to rxe_mr_cleanup to allow operations
holding a reference to the mr to complete.
Link: https://lore.kernel.org/r/20220421014042.26985-6-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Move the code from rxe_qp_destroy() to rxe_qp_do_cleanup(). This allows
flows holding references to qp to complete before the qp object is torn
down.
Link: https://lore.kernel.org/r/20220421014042.26985-5-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
In the tasklets (completer, responder, and requester) check the return
value from rxe_get() to detect failures to get a reference. This only
occurs if the qp has had its reference count drop to zero which indicates
that it no longer should be used.
The ref is never 0 today because the tasklets are flushed before the ref
is dropped. The next patch changes this so that the ref is dropped then
the tasklets are flushed.
Link: https://lore.kernel.org/r/20220421014042.26985-4-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Move cleanup code from rxe_destroy_srq() to rxe_srq_cleanup() which is
called after all references are dropped to allow code depending on the srq
object to complete.
Link: https://lore.kernel.org/r/20220421014042.26985-3-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Currently the #define IB_SRQ_INIT_MASK is used to distinguish the
rxe_create_srq verb from the rxe_modify_srq verb so that some code can be
shared between these two subroutines.
This commit splits rxe_srq_chk_attr into two subroutines: rxe_srq_chk_init
and rxe_srq_chk_attr which handle the create_srq and modify_srq verbs
separately.
Link: https://lore.kernel.org/r/20220421014042.26985-2-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
For write request the remote addr will be sent only with first packet so
we don't have to adjust wqe->iova in retry operation.
Link: https://lore.kernel.org/r/20220502053907.6388-1-cgxu519@mykernel.net
Signed-off-by: Chengguang Xu <cgxu519@mykernel.net>
Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Based on the commit c9f4c69583 ("RDMA/rxe: Reverse the sense of
RXE_POOL_NO_ALLOC"), only the mr pool uses the RXE_POOL_ALLOC, As such,
replace this flags with pool type to save memory.
Link: https://lore.kernel.org/r/20220428041028.1363139-1-yanjun.zhu@linux.dev
Signed-off-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
rxe_mcast.c currently uses _irqsave spinlocks for rxe->mcg_lock while
rxe_recv.c uses _bh spinlocks for the same lock.
As there is no case where the mcg_lock can be taken from an IRQ, change
these all to bh locks so we don't have confusing mismatched lock types on
the same spinlock.
Fixes: 6090a0c4c7 ("RDMA/rxe: Cleanup rxe_mcast.c")
Link: https://lore.kernel.org/r/20220504202817.98247-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
These routines were not intended to be called under a spinlock and will
throw debugging warnings:
raw_local_irq_restore() called with IRQs enabled
WARNING: CPU: 13 PID: 3107 at kernel/locking/irqflag-debug.c:10 warn_bogus_irq_restore+0x2f/0x50
CPU: 13 PID: 3107 Comm: python3 Tainted: G E 5.18.0-rc1+ #7
Hardware name: innotek GmbH VirtualBox/VirtualBox, BIOS VirtualBox 12/01/2006
RIP: 0010:warn_bogus_irq_restore+0x2f/0x50
Call Trace:
<TASK>
_raw_spin_unlock_irqrestore+0x75/0x80
rxe_attach_mcast+0x304/0x480 [rdma_rxe]
ib_attach_mcast+0x88/0xa0 [ib_core]
ib_uverbs_attach_mcast+0x186/0x1e0 [ib_uverbs]
ib_uverbs_handler_UVERBS_METHOD_INVOKE_WRITE+0xcd/0x140 [ib_uverbs]
ib_uverbs_cmd_verbs+0xdb0/0xea0 [ib_uverbs]
ib_uverbs_ioctl+0xd2/0x160 [ib_uverbs]
do_syscall_64+0x5c/0x80
entry_SYSCALL_64_after_hwframe+0x44/0xae
Move them out of the spinlock, it is OK if there is some races setting up
the MC reception at the ethernet layer with rbtree lookups.
Fixes: 6090a0c4c7 ("RDMA/rxe: Cleanup rxe_mcast.c")
Link: https://lore.kernel.org/r/20220504202817.98247-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The calling of siw_cm_upcall and detaching new_cep with its listen_cep
should be atomistic semantics. Otherwise siw_reject may be called in a
temporary state, e,g, siw_cm_upcall is called but the new_cep->listen_cep
has not being cleared.
This fixes a WARN:
WARNING: CPU: 7 PID: 201 at drivers/infiniband/sw/siw/siw_cm.c:255 siw_cep_put+0x125/0x130 [siw]
CPU: 2 PID: 201 Comm: kworker/u16:22 Kdump: loaded Tainted: G E 5.17.0-rc7 #1
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014
Workqueue: iw_cm_wq cm_work_handler [iw_cm]
RIP: 0010:siw_cep_put+0x125/0x130 [siw]
Call Trace:
<TASK>
siw_reject+0xac/0x180 [siw]
iw_cm_reject+0x68/0xc0 [iw_cm]
cm_work_handler+0x59d/0xe20 [iw_cm]
process_one_work+0x1e2/0x3b0
worker_thread+0x50/0x3a0
? rescuer_thread+0x390/0x390
kthread+0xe5/0x110
? kthread_complete_and_exit+0x20/0x20
ret_from_fork+0x1f/0x30
</TASK>
Fixes: 6c52fdc244 ("rdma/siw: connection management")
Link: https://lore.kernel.org/r/d528d83466c44687f3872eadcb8c184528b2e2d4.1650526554.git.chengyou@linux.alibaba.com
Reported-by: Luis Chamberlain <mcgrof@kernel.org>
Reviewed-by: Bernard Metzler <bmt@zurich.ibm.com>
Signed-off-by: Cheng Xu <chengyou@linux.alibaba.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
In finish_packet() in rxe_req.c a variable was incorrectly called paylen
instead of payload. Elsewhere in the rxe source payload is always used for
the RoCE payload length and paylen is always used for the UDP payload
length. This will cause unnecessary confusion.
Replace paylen by payload in finish_packet().
Link: https://lore.kernel.org/r/20220420172316.5465-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The rping benchmark fails on long runs. The root cause of this failure has
been traced to a failure to compute a nonzero value of mr in rare
situations.
Fix this failure by correctly handling the computation of mr in
read_reply() in rxe_resp.c in the replay flow.
Fixes: 8a1a0be894 ("RDMA/rxe: Replace mr by rkey in responder resources")
Link: https://lore.kernel.org/r/20220418174103.3040-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The referenced commit generates a reference counting error if the rkey has
the same index but the wrong key. In this case the reference taken by
rxe_pool_get_index() is not dropped.
Drop the reference if the keys don't match in rxe_recheck_mr(). Check
that the mw and mr are still valid.
Fixes: 8a1a0be894 ("RDMA/rxe: Replace mr by rkey in responder resources")
Link: https://lore.kernel.org/r/20220411030647.20011-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Current rxe_requester() doesn't generate a completion when processing an
unsupported/invalid opcode. If rxe driver doesn't support a new opcode
(e.g. RDMA Atomic Write) and RDMA library supports it, an application
using the new opcode can reproduce this issue. Fix the issue by calling
"goto err;".
Fixes: 8700e3e7c4 ("Soft RoCE driver")
Link: https://lore.kernel.org/r/20220410113513.27537-1-yangx.jy@fujitsu.com
Signed-off-by: Xiao Yang <yangx.jy@fujitsu.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The rdma_rxe driver does not actually support the reliable datagram
transport but contains two references to RD opcodes in driver code. This
commit removes these references to RD transport opcodes which are never
used.
Link: https://lore.kernel.org/r/cce0f07d-25fc-5880-69e7-001d951750b7@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Currently the rdma_rxe driver supports SMI type QPs in a few places which
is incorrect. RoCE devices never should support SMI QPs. This commit
removes SMI QP support from the driver.
Link: https://lore.kernel.org/r/20220407185416.16372-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Currently the rdma_rxe driver claims to support both 2A and 2B type memory
windows. But the IBA requires
010-37.2.31: If an HCA supports the Base Memory Management
extensions, the HCA shall support either Type 2A or Type 2B MWs,
but not both.
This commit removes the device capability bit for type 2A memory windows
and adds a clarifying comment to rxe_mw.c.
Link: https://lore.kernel.org/r/20220407184321.14207-1-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Split out flags from ib_device::device_cap_flags that are only used
internally to the kernel into kernel_cap_flags that is not part of the
uapi. This limits the device_cap_flags to being the same bitmap that will
be copied to userspace.
This cleanly splits out the uverbs flags from the kernel flags to avoid
confusion in the flags bitmap.
Add some short comments describing which each of the kernel flags is
connected to. Remove unused kernel flags.
Link: https://lore.kernel.org/r/0-v2-22c19e565eef+139a-kern_caps_jgg@nvidia.com
Reviewed-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Max Gurtovoy <mgurtovoy@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The documentation of the function rvt_error_qp says both r_lock and
s_lock need to be held when calling that function.
It also asserts using lockdep that both of those locks are held.
rvt_error_qp is called form rvt_send_cq, which is called from
rvt_qp_complete_swqe, which is called from rvt_send_complete, which is
called from rvt_ruc_loopback in two places. Both of these places do not
hold r_lock. Fix this by acquiring a spin_lock of r_lock in both of
these places.
The r_lock acquiring cannot be added in rvt_qp_complete_swqe because
some of its other callers already have r_lock acquired.
Link: https://lore.kernel.org/r/20220228195144.71946-1-dossche.niels@gmail.com
Signed-off-by: Niels Dossche <dossche.niels@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The documentation of the function rvt_error_qp says both r_lock and s_lock
need to be held when calling that function. It also asserts using lockdep
that both of those locks are held. However, the commit I referenced in
Fixes accidentally makes the call to rvt_error_qp in rvt_ruc_loopback no
longer covered by r_lock. This results in the lockdep assertion failing
and also possibly in a race condition.
Fixes: d757c60eca ("IB/rdmavt: Fix concurrency panics in QP post_send and modify to error")
Link: https://lore.kernel.org/r/20220228165330.41546-1-dossche.niels@gmail.com
Signed-off-by: Niels Dossche <dossche.niels@gmail.com>
Acked-by: Dennis Dalessandro <dennis.dalessandro@cornelisnetworks.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
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()
...
Rename rxe_add_ref() to rxe_get() and rxe_drop_ref() to rxe_put().
Significantly improves readability for new readers.
Link: https://lore.kernel.org/r/20220304000808.225811-10-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Currently the rxe driver uses red-black trees to add indices to the rxe
object pools. Linux xarrays provide a better way to implement the same
functionality for indices. This patch replaces red-black trees by xarrays
for pool objects. Since xarrays already have a spinlock use that in place
of the pool rwlock. Make sure that all changes in the xarray(index) and
kref(ref counnt) occur atomically.
Link: https://lore.kernel.org/r/20220304000808.225811-9-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Move the maximum number of elements from a parameter in rxe_pool_init to a
member of the rxe_type_info array.
Link: https://lore.kernel.org/r/20220304000808.225811-7-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
There is only one remaining object type that allocates its own memory,
that is mr. So the sense of RXE_POOL_NO_ALLOC is changed to
RXE_POOL_ALLOC. Add checks to rxe_alloc() and rxe_add_to_pool() to make
sure the correct call is used for the setting of this flag.
Link: https://lore.kernel.org/r/20220304000808.225811-4-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Currently rxe saves a copy of MR in responder resources for RDMA reads.
Since the responder resources are never freed just over written if more
are needed this MR may not have a reference freed until the QP is
destroyed. This patch uses the rkey instead of the MR and on subsequent
packets of a multipacket read reply message it looks up the MR from the
rkey for each packet. This makes it possible for a user to deregister an
MR or unbind a MW on the fly and get correct behaviour.
Link: https://lore.kernel.org/r/20220304000808.225811-3-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The commit referenced below can take a reference to the AH which is never
dropped. This only happens in the UD request path. This patch optionally
passes that AH back to the caller so that it can hold the reference while
the AV is being accessed and then drop it. Code to do this is added to
rxe_req.c. The AV is also passed to rxe_prepare in rxe_net.c as an
optimization.
Fixes: e2fe06c908 ("RDMA/rxe: Lookup kernel AH from ah index in UD WQEs")
Link: https://lore.kernel.org/r/20220304000808.225811-2-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The argument 'payload' is not used in update_state(), so just remove it.
Link: https://lore.kernel.org/r/20220307145047.3235675-2-cgxu519@mykernel.net
Signed-off-by: Chengguang Xu <cgxu519@mykernel.net>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Acked-by: Zhu Yanjun <zyjzyj2000@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
The type of wqe length is u32 so in order to avoid overflow and shadow
casting change variable and relevant function argument to proper type.
Link: https://lore.kernel.org/r/20220307145047.3235675-1-cgxu519@mykernel.net
Signed-off-by: Chengguang Xu <cgxu519@mykernel.net>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
