| Age | Commit message (Collapse) | Author |
|---|
|
Implement the two easy cases of ib_rereg_user_mr.
Link: https://lore.kernel.org/r/20230530221334.89432-7-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
In order to conform to other drivers stop using rkey == 0 as an indication
that there are no remote access flags set. Set rkey == lkey by default
for all MRs.
Link: https://lore.kernel.org/r/20230530221334.89432-6-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
Introduce supported bit masks for setting the access attributes of MWs,
MRs, and QPs. Check these when attributes are set.
Link: https://lore.kernel.org/r/20230530221334.89432-5-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
The subroutine rxe_check_bind_mw() in rxe_mw.c performs checks on the mw
access flags before they are set so they always succeed. This patch
instead checks the access flags passed in the send wqe.
Fixes: 32a577b4c3a9 ("RDMA/rxe: Add support for bind MW work requests")
Link: https://lore.kernel.org/r/20230530221334.89432-4-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
Bypass calling check_rkey() in rxe_resp.c for non-rdma messages.
Link: https://lore.kernel.org/r/20230530221334.89432-3-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
Rename IB_ACCESS_REMOTE to RXE_ACCESS_REMOTE and move to rxe_verbs.h as an
enum instead of a #define. Shouldn't use IB_xxx for rxe symbols.
Link: https://lore.kernel.org/r/20230530221334.89432-2-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
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: cdd0b85675ae ("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: 8700e3e7c485 ("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: 78b26a335310 ("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>
|
|
The commit 9b4b7c1f9f54 ("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: 9b4b7c1f9f54 ("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>
|
|
spin_{lock_irqsave,unlock_irqrestore}
We need to call spin_lock_irqsave()/spin_unlock_irqrestore() for
state_lock in rxe, otherwsie the callchain:
ib_post_send_mad
-> spin_lock_irqsave
-> ib_post_send -> rxe_post_send
-> spin_lock_bh
-> spin_unlock_bh
-> spin_unlock_irqrestore
Causes below traces during run block nvmeof-mp/001 test due to mismatched
spinlock nesting:
WARNING: CPU: 0 PID: 94794 at kernel/softirq.c:376 __local_bh_enable_ip+0xc2/0x140
[ ... ]
CPU: 0 PID: 94794 Comm: kworker/u4:1 Tainted: G E 6.4.0-rc1 #9
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.15.0-0-g2dd4b9b-rebuilt.opensuse.org 04/01/2014
Workqueue: rdma_cm cma_work_handler [rdma_cm]
RIP: 0010:__local_bh_enable_ip+0xc2/0x140
Code: 48 85 c0 74 72 5b 41 5c 5d 31 c0 89 c2 89 c1 89 c6 89 c7 41 89 c0 e9 bd 0e 11 01 65 8b 05 f2 65 72 48 85 c0 0f 85 76 ff ff ff <0f> 0b e9 6f ff ff ff e8 d2 39 1c 00 eb 80 4c 89 e7 e8 68 ad 0a 00
RSP: 0018:ffffb7cf818539f0 EFLAGS: 00010046
RAX: 0000000000000000 RBX: 0000000000000201 RCX: 0000000000000000
RDX: 0000000000000000 RSI: 0000000000000201 RDI: ffffffffc0f25f79
RBP: ffffb7cf81853a00 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000000 R12: ffffffffc0f25f79
R13: ffff8db1f0fa6000 R14: ffff8db2c63ff000 R15: 00000000000000e8
FS: 0000000000000000(0000) GS:ffff8db33bc00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000559758db0f20 CR3: 0000000105124000 CR4: 00000000003506f0
Call Trace:
<TASK>
_raw_spin_unlock_bh+0x31/0x40
rxe_post_send+0x59/0x8b0 [rdma_rxe]
ib_send_mad+0x26b/0x470 [ib_core]
ib_post_send_mad+0x150/0xb40 [ib_core]
? cm_form_tid+0x5b/0x90 [ib_cm]
ib_send_cm_req+0x7c8/0xb70 [ib_cm]
rdma_connect_locked+0x433/0x940 [rdma_cm]
nvme_rdma_cm_handler+0x5d7/0x9c0 [nvme_rdma]
cma_cm_event_handler+0x4f/0x170 [rdma_cm]
cma_work_handler+0x6a/0xe0 [rdma_cm]
process_one_work+0x2a9/0x580
worker_thread+0x52/0x3f0
? __pfx_worker_thread+0x10/0x10
kthread+0x109/0x140
? __pfx_kthread+0x10/0x10
ret_from_fork+0x2c/0x50
</TASK>
raw_local_irq_restore() called with IRQs enabled
WARNING: CPU: 0 PID: 94794 at kernel/locking/irqflag-debug.c:10 warn_bogus_irq_restore+0x37/0x60
[ ... ]
CPU: 0 PID: 94794 Comm: kworker/u4:1 Tainted: G W E 6.4.0-rc1 #9
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.15.0-0-g2dd4b9b-rebuilt.opensuse.org 04/01/2014
Workqueue: rdma_cm cma_work_handler [rdma_cm]
RIP: 0010:warn_bogus_irq_restore+0x37/0x60
Code: fb 01 77 36 83 e3 01 74 0e 48 8b 5d f8 c9 31 f6 89 f7 e9 ac ea 01 00 48 c7 c7 e0 52 33 b9 c6 05 bb 1c 69 01 01 e8 39 24 f0 fe <0f> 0b 48 8b 5d f8 c9 31 f6 89 f7 e9 89 ea 01 00 0f b6 f3 48 c7 c7
RSP: 0018:ffffb7cf81853a58 EFLAGS: 00010246
RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000000
RBP: ffffb7cf81853a60 R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000000 R12: ffff8db2cfb1a9e8
R13: ffff8db2cfb1a9d8 R14: ffff8db2c63ff000 R15: 0000000000000000
FS: 0000000000000000(0000) GS:ffff8db33bc00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000559758db0f20 CR3: 0000000105124000 CR4: 00000000003506f0
Call Trace:
<TASK>
_raw_spin_unlock_irqrestore+0x91/0xa0
ib_send_mad+0x1e3/0x470 [ib_core]
ib_post_send_mad+0x150/0xb40 [ib_core]
? cm_form_tid+0x5b/0x90 [ib_cm]
ib_send_cm_req+0x7c8/0xb70 [ib_cm]
rdma_connect_locked+0x433/0x940 [rdma_cm]
nvme_rdma_cm_handler+0x5d7/0x9c0 [nvme_rdma]
cma_cm_event_handler+0x4f/0x170 [rdma_cm]
cma_work_handler+0x6a/0xe0 [rdma_cm]
process_one_work+0x2a9/0x580
worker_thread+0x52/0x3f0
? __pfx_worker_thread+0x10/0x10
kthread+0x109/0x140
? __pfx_kthread+0x10/0x10
ret_from_fork+0x2c/0x50
</TASK>
Fixes: f605f26ea196 ("RDMA/rxe: Protect QP state with qp->state_lock")
Link: https://lore.kernel.org/r/20230510035056.881196-1-guoqing.jiang@linux.dev
Signed-off-by: Guoqing Jiang <guoqing.jiang@linux.dev>
Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
A recent patch can cause a double spin_unlock_bh() in rxe_qp_to_attr() at
line 715 in rxe_qp.c. Move the 2nd unlock into the if statement.
Fixes: f605f26ea196 ("RDMA/rxe: Protect QP state with qp->state_lock")
Link: https://lore.kernel.org/r/20230515201056.1591140-1-rpearsonhpe@gmail.com
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Closes: https://lore.kernel.org/r/27773078-40ce-414f-8b97-781954da9f25@kili.mountain
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
The following deadlock is observed:
Call Trace:
<IRQ>
_raw_spin_lock_bh+0x29/0x30
check_type_state.constprop.0+0x4e/0xc0 [rdma_rxe]
rxe_rcv+0x173/0x3d0 [rdma_rxe]
rxe_udp_encap_recv+0x69/0xd0 [rdma_rxe]
? __pfx_rxe_udp_encap_recv+0x10/0x10 [rdma_rxe]
udp_queue_rcv_one_skb+0x258/0x520
udp_unicast_rcv_skb+0x75/0x90
__udp4_lib_rcv+0x364/0x5c0
ip_protocol_deliver_rcu+0xa7/0x160
ip_local_deliver_finish+0x73/0xa0
ip_sublist_rcv_finish+0x80/0x90
ip_sublist_rcv+0x191/0x220
ip_list_rcv+0x132/0x160
__netif_receive_skb_list_core+0x297/0x2c0
netif_receive_skb_list_internal+0x1c5/0x300
napi_complete_done+0x6f/0x1b0
virtnet_poll+0x1f4/0x2d0 [virtio_net]
__napi_poll+0x2c/0x1b0
net_rx_action+0x293/0x350
? __napi_schedule+0x79/0x90
__do_softirq+0xcb/0x2ab
__irq_exit_rcu+0xb9/0xf0
common_interrupt+0x80/0xa0
</IRQ>
<TASK>
asm_common_interrupt+0x22/0x40
RIP: 0010:_raw_spin_lock+0x17/0x30
rxe_requester+0xe4/0x8f0 [rdma_rxe]
? xas_load+0x9/0xa0
? xa_load+0x70/0xb0
do_task+0x64/0x1f0 [rdma_rxe]
rxe_post_send+0x54/0x110 [rdma_rxe]
ib_uverbs_post_send+0x5f8/0x680 [ib_uverbs]
? netif_receive_skb_list_internal+0x1e3/0x300
ib_uverbs_write+0x3c8/0x500 [ib_uverbs]
vfs_write+0xc5/0x3b0
ksys_write+0xab/0xe0
? syscall_trace_enter.constprop.0+0x126/0x1a0
do_syscall_64+0x3b/0x90
entry_SYSCALL_64_after_hwframe+0x72/0xdc
</TASK>
The deadlock is easily reproducible with perftest. Fix it by disabling
softirq when acquiring the lock in process context.
Fixes: f605f26ea196 ("RDMA/rxe: Protect QP state with qp->state_lock")
Link: https://lore.kernel.org/r/20230418090642.1849358-1-matsuda-daisuke@fujitsu.com
Signed-off-by: Daisuke Matsuda <matsuda-daisuke@fujitsu.com>
Acked-by: Zhu Yanjun <zyjzyj2000@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
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>
|
|
In the function rxe_create_qp(), rxe_qp_from_init() is called to
initialize qp, internally things like rxe_init_task are not setup until
rxe_qp_init_req().
If an error occurred 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.
If rxe_init_task is not executed, rxe_cleanup_task will not be called.
Reported-by: syzbot+cfcc1a3c85be15a40cba@syzkaller.appspotmail.com
Link: https://syzkaller.appspot.com/bug?id=fd85757b74b3eb59f904138486f755f71e090df8
Fixes: 8700e3e7c485 ("Soft RoCE driver")
Fixes: 2d4b21e0a291 ("IB/rxe: Prevent from completer to operate on non valid QP")
Signed-off-by: Zhu Yanjun <yanjun.zhu@linux.dev>
Link: https://lore.kernel.org/r/20230413101115.1366068-1-yanjun.zhu@intel.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
|
|
In a previous patch TASKLET_STATE_SCHED was used as a mask but it is a bit
position instead. Add the missing shift.
Link: https://lore.kernel.org/r/20230329193308.7489-1-rpearsonhpe@gmail.com
Reported-by: Dan Carpenter <error27@gmail.com>
Link: https://lore.kernel.org/linux-rdma/8a054b78-6d50-4bc6-8d8a-83f85fbdb82f@kili.mountain/
Fixes: d94671632572 ("RDMA/rxe: Rewrite rxe_task.c")
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
There is no need to print any debug messages after failure to
allocate memory, because kernel will print OOM dumps anyway.
Together with removal of these messages, remove useless goto jumps.
Fixes: 5bf944f24129 ("RDMA/rxe: Add error messages")
Reported-by: Dan Carpenter <error27@gmail.com>
Link: https://lore.kernel.org/all/ea43486f-43dd-4054-b1d5-3a0d202be621@kili.mountain
Link: https://lore.kernel.org/r/d3cedf723b84e73e8062a67b7489d33802bafba2.1680113597.git.leon@kernel.org
Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
|
|
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: 8700e3e7c485 ("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: c6aba5ea0055 ("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>
|
|
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: 0a67c46d2e99 ("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: 8700e3e7c485 ("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>
|
|
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>
|
|
Move rxe_map_mr_sg() to rxe_mr.c where it makes a little more sense.
Link: https://lore.kernel.org/r/20230119235936.19728-3-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
Remove blank lines and replace EFAULT by EINVAL when an invalid
mr type is used.
Link: https://lore.kernel.org/r/20230119235936.19728-2-rpearsonhpe@gmail.com
Signed-off-by: Bob Pearson <rpearsonhpe@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
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: 0994a1bcd5f7 ("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: 3ccffe8abf2f ("RDMA/rxe: Move max_elem into rxe_type_info")
Fixes: 3225717f6dfa ("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>
|
|
Move the conditional code into a function, with two varients so it is
harder to make these kinds of mistakes.
drivers/infiniband/sw/rxe/rxe_resp.c: In function 'atomic_write_reply':
drivers/infiniband/sw/rxe/rxe_resp.c:794:13: error: unused variable 'payload' [-Werror=unused-variable]
794 | int payload = payload_size(pkt);
| ^~~~~~~
drivers/infiniband/sw/rxe/rxe_resp.c:793:24: error: unused variable 'mr' [-Werror=unused-variable]
793 | struct rxe_mr *mr = qp->resp.mr;
| ^~
drivers/infiniband/sw/rxe/rxe_resp.c:791:19: error: unused variable 'dst' [-Werror=unused-variable]
791 | u64 src, *dst;
| ^~~
drivers/infiniband/sw/rxe/rxe_resp.c:791:13: error: unused variable 'src' [-Werror=unused-variable]
791 | u64 src, *dst;
Fixes: 034e285f8b99 ("RDMA/rxe: Make responder support atomic write on RC service")
Link: https://lore.kernel.org/linux-rdma/Y5s+EVE7eLWQqOwv@nvidia.com/
Reported-by: Guenter Roeck <linux@roeck-us.net>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
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>
|
