Age | Commit message (Collapse) | Author |
|
[ Upstream commit e3ddd6067ee62f6e76ebcf61ff08b2c729ae412b ]
When page_address() fails, umem should be freed just like when
rxe_mem_alloc() fails.
Fixes: 8700e3e7c485 ("Soft RoCE driver")
Link: https://lore.kernel.org/r/20200819075632.22285-1-dinghao.liu@zju.edu.cn
Signed-off-by: Dinghao Liu <dinghao.liu@zju.edu.cn>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit baa5b28b7a474f66a511ebf71a2ade510652a2f6 ]
There are error cases when we will call free_srv before device kobject is
initialized; in such cases calling put_device generates the following
warning:
kobject: '(null)' (000000009f5445ed): is not initialized, yet
kobject_put() is being called.
So call device_initialize() only once when the server is allocated. If we
end up calling put_srv() and subsequently free_srv(), our call to
put_device() would result in deletion of the obj. Call device_add() later
when we actually have a connection. Correspondingly, call device_del()
instead of device_unregister() when srv->dev_ref falls to 0.
Fixes: 9cb837480424 ("RDMA/rtrs: server: main functionality")
Link: https://lore.kernel.org/r/20200811092722.2450-1-haris.iqbal@cloud.ionos.com
Suggested-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Md Haris Iqbal <haris.iqbal@cloud.ionos.com>
Reviewed-by: Jack Wang <jinpu.wang@cloud.ionos.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit f6a9d47ae6854980fc4b1676f1fe9f9fa45ea4e2 ]
When a rdma_cm_id needs to be destroyed after a handler callback fails,
part of the destruction pattern is open coded into each call site.
Unfortunately the blind assignment to state discards important information
needed to do cma_cancel_operation(). This results in active operations
being left running after rdma_destroy_id() completes, and the
use-after-free bugs from KASAN.
Consolidate this entire pattern into destroy_id_handler_unlock() and
manage the locking correctly. The state should be set to
RDMA_CM_DESTROYING under the handler_lock to atomically ensure no futher
handlers are called.
Link: https://lore.kernel.org/r/20200723070707.1771101-5-leon@kernel.org
Reported-by: syzbot+08092148130652a6faae@syzkaller.appspotmail.com
Reported-by: syzbot+a929647172775e335941@syzkaller.appspotmail.com
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit cc9c037343898eb7a775e6b81d092ee21eeff218 ]
The REQ flows are concerned that once the handler is called on the new
cm_id the ULP can choose to trigger a rdma_destroy_id() concurrently at
any time.
However, this is not true, while the ULP can call rdma_destroy_id(), it
immediately blocks on the handler_mutex which prevents anything harmful
from running concurrently.
Remove the confusing extra locking and refcounts and make the
handler_mutex protecting state during destroy more clear.
Link: https://lore.kernel.org/r/20200723070707.1771101-4-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 3647a28de1ada8708efc78d956619b9df5004478 ]
Whenever an event is delivered to the handler it should be done under the
handler_mutex and upon any non-zero return from the handler it should
trigger destruction of the cm_id.
cma_process_remove() skips some steps here, it is not necessarily wrong
since the state change should prevent any races, but it is confusing and
unnecessary.
Follow the standard pattern here, with the slight twist that the
transition to RDMA_CM_DEVICE_REMOVAL includes a cma_cancel_operation().
Link: https://lore.kernel.org/r/20200723070707.1771101-3-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit d54f23c09ec62670901f1a2a4712a5218522ca2b ]
cma_process_remove() triggers an unconditional rdma_destroy_id() for
internal_id's and skips the event deliver and transition through
RDMA_CM_DEVICE_REMOVAL.
This is confusing and unnecessary. internal_id always has
cma_listen_handler() as the handler, have it catch the
RDMA_CM_DEVICE_REMOVAL event and directly consume it and signal removal.
This way the FSM sequence never skips the DEVICE_REMOVAL case and the
logic in this hard to test area is simplified.
Link: https://lore.kernel.org/r/20200723070707.1771101-2-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 6da06c6291f38be4df6df2efb76ba925096d2691 ]
This patch caused some issues on SEND operation, and it should be reverted
to make the drivers work correctly. There will be a better solution that
has been tested carefully to solve the original problem.
This reverts commit 711195e57d341e58133d92cf8aaab1db24e4768d.
Fixes: 711195e57d34 ("RDMA/hns: Reserve one sge in order to avoid local length error")
Link: https://lore.kernel.org/r/1597829984-20223-1-git-send-email-liweihang@huawei.com
Signed-off-by: Weihang Li <liweihang@huawei.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit a812f2d60a9fb7818f9c81f967180317b52545c0 ]
Driver shall add only the kernel qps to the flush list for clean up.
During async error events from the HW, driver is adding qps to this list
without checking if the qp is kernel qp or not.
Add a check to avoid user qp addition to the flush list.
Fixes: 942c9b6ca8de ("RDMA/bnxt_re: Avoid Hard lockup during error CQE processing")
Fixes: c50866e2853a ("bnxt_re: fix the regression due to changes in alloc_pbl")
Link: https://lore.kernel.org/r/1596689148-4023-1-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
commit b25e8e85e75a61af1ddc88c4798387dd3132dd43 upstream.
The following message occurs when running an AI application with TID RDMA
enabled:
hfi1 0000:7f:00.0: hfi1_0: [QP74] hfi1_tid_timeout 4084
hfi1 0000:7f:00.0: hfi1_0: [QP70] hfi1_tid_timeout 4084
The issue happens when TID RDMA WRITE request is followed by an
IB_WR_RDMA_WRITE_WITH_IMM request, the latter could be completed first on
the responder side. As a result, no ACK packet for the latter could be
sent because the TID RDMA WRITE request is still being processed on the
responder side.
When the TID RDMA WRITE request is eventually completed, the requester
will wait for the IB_WR_RDMA_WRITE_WITH_IMM request to be acknowledged.
If the next request is another TID RDMA WRITE request, no TID RDMA WRITE
DATA packet could be sent because the preceding IB_WR_RDMA_WRITE_WITH_IMM
request is not completed yet.
Consequently the IB_WR_RDMA_WRITE_WITH_IMM will be retried but it will be
ignored on the responder side because the responder thinks it has already
been completed. Eventually the retry will be exhausted and the qp will be
put into error state on the requester side. On the responder side, the TID
resource timer will eventually expire because no TID RDMA WRITE DATA
packets will be received for the second TID RDMA WRITE request. There is
also risk of a write-after-write memory corruption due to the issue.
Fix by adding a requester side interlock to prevent any potential data
corruption and TID RDMA protocol error.
Fixes: a0b34f75ec20 ("IB/hfi1: Add interlock between a TID RDMA request and other requests")
Link: https://lore.kernel.org/r/20200811174931.191210.84093.stgit@awfm-01.aw.intel.com
Cc: <stable@vger.kernel.org> # 5.4.x+
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Reviewed-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
|
|
[ Upstream commit cbeb7d896c0f296451ffa7b67e7706786b8364c8 ]
In manual mode allow bind user QPs with different pids to same counter,
since this is allowed in auto mode.
Bind kernel QPs and user QPs to the same counter are not allowed.
Fixes: 1bd8e0a9d0fd ("RDMA/counter: Allow manual mode configuration support")
Link: https://lore.kernel.org/r/20200702082933.424537-4-leon@kernel.org
Signed-off-by: Mark Zhang <markz@mellanox.com>
Reviewed-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit c9f557421e505f75da4234a6af8eff46bc08614b ]
In auto mode only bind user QPs to a dynamic counter, since this feature
is mainly used for system statistic and diagnostic purpose, while there's
no need to counter kernel QPs so far.
Fixes: 99fa331dc862 ("RDMA/counter: Add "auto" configuration mode support")
Link: https://lore.kernel.org/r/20200702082933.424537-3-leon@kernel.org
Signed-off-by: Mark Zhang <markz@mellanox.com>
Reviewed-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 04c0a5fcfcf65aade2fb238b6336445f1a99b646 ]
Set IOVA on IB MR in uverbs layer to let all drivers have it, this
includes both reg/rereg MR flows.
As part of this change cleaned-up this setting from the drivers that
already did it by themselves in their user flows.
Fixes: e6f0330106f4 ("mlx4_ib: set user mr attributes in struct ib_mr")
Link: https://lore.kernel.org/r/20200630093916.332097-3-leon@kernel.org
Signed-off-by: Yishai Hadas <yishaih@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 65936bf25f90fe440bb2d11624c7d10fab266639 ]
ipoib_mcast_carrier_on_task() insanely open codes a rtnl_lock() such that
the only time flush_workqueue() can be called is if it also clears
IPOIB_FLAG_OPER_UP.
Thus the flush inside ipoib_flush_ah() will deadlock if it gets unlucky
enough, and lockdep doesn't help us to find it early:
CPU0 CPU1 CPU2
__ipoib_ib_dev_flush()
down_read(vlan_rwsem)
ipoib_vlan_add()
rtnl_trylock()
down_write(vlan_rwsem)
ipoib_mcast_carrier_on_task()
while (!rtnl_trylock())
msleep(20);
ipoib_flush_ah()
flush_workqueue(priv->wq)
Clean up the ah_reaper related functions and lifecycle to make sense:
- Start/Stop of the reaper should only be done in open/stop NDOs, not in
any other places
- cancel and flush of the reaper should only happen in the stop NDO.
cancel is only functional when combined with IPOIB_STOP_REAPER.
- Non-stop places were flushing the AH's just need to flush out dead AH's
synchronously and ignore the background task completely. It is fully
locked and harmless to leave running.
Which ultimately fixes the ABBA deadlock by removing the unnecessary
flush_workqueue() from the problematic place under the vlan_rwsem.
Fixes: efc82eeeae4e ("IB/ipoib: No longer use flush as a parameter")
Link: https://lore.kernel.org/r/20200625174219.290842-1-kamalheib1@gmail.com
Reported-by: Kamal Heib <kheib@redhat.com>
Tested-by: Kamal Heib <kheib@redhat.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 95a5631f6c9f3045f26245e6045244652204dfdb ]
The return value from ipoib_ib_dev_stop() is always 0 - change it to be
void.
Link: https://lore.kernel.org/r/20200623105236.18683-1-kamalheib1@gmail.com
Signed-off-by: Kamal Heib <kamalheib1@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 395f2e8fd340c5bfad026f5968b56ec34cf20dd1 ]
If the hns ROCEE reports a general error CQE (types not specified by the IB
General Specifications), it's no need to change the QP state to error, and
the driver should just skip it.
Fixes: 7c044adca272 ("RDMA/hns: Simplify the cqe code of poll cq")
Link: https://lore.kernel.org/r/1595932941-40613-8-git-send-email-liweihang@huawei.com
Signed-off-by: Xi Wang <wangxi11@huawei.com>
Signed-off-by: Weihang Li <liweihang@huawei.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 4327bd2c41412657ee2c8c0d8d3d1945268b4238 ]
One qp state migrations legal configuration was deleted mistakenly.
Fixes: 357f34294686 ("RDMA/hns: Simplify the state judgment code of qp")
Link: https://lore.kernel.org/r/1595932941-40613-7-git-send-email-liweihang@huawei.com
Signed-off-by: Lang Cheng <chenglang@huawei.com>
Signed-off-by: Weihang Li <liweihang@huawei.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 1d70ad0f85435a7262de802b104e49e6598c50ff ]
When dumping QPs bound to a counter, raw QPs should be allowed to dump
without the CAP_NET_RAW privilege. This is consistent with what "rdma res
show qp" does.
Fixes: c4ffee7c9bdb ("RDMA/netlink: Implement counter dumpit calback")
Link: https://lore.kernel.org/r/20200727095828.496195-1-leon@kernel.org
Signed-off-by: Mark Zhang <markz@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 03ed5a8cda659e3c71d106b0dd4ce6520e4dcd6e ]
lockdep triggers a warning from time to time when running a regression
test:
rnbd_client L685: </dev/nullb0@bla> Device disconnected.
rnbd_client L1756: Unloading module
workqueue: WQ_MEM_RECLAIM rtrs_client_wq:rtrs_clt_reconnect_work [rtrs_client] is flushing !WQ_MEM_RECLAIM ib_addr:process_one_req [ib_core]
WARNING: CPU: 2 PID: 18824 at kernel/workqueue.c:2517 check_flush_dependency+0xad/0x130
The root cause is workqueue core expect flushing should not be done for a
!WQ_MEM_RECLAIM wq from a WQ_MEM_RECLAIM workqueue.
In above case ib_addr workqueue without WQ_MEM_RECLAIM, but rtrs_wq
WQ_MEM_RECLAIM.
To avoid the warning, remove the WQ_MEM_RECLAIM flag.
Fixes: 9cb837480424 ("RDMA/rtrs: server: main functionality")
Fixes: 6a98d71daea1 ("RDMA/rtrs: client: main functionality")
Link: https://lore.kernel.org/r/20200724111508.15734-4-haris.iqbal@cloud.ionos.com
Signed-off-by: Jack Wang <jinpu.wang@cloud.ionos.com>
Signed-off-by: Md Haris Iqbal <haris.iqbal@cloud.ionos.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 09e0dbbeed82e35ce2cd21e086a6fac934163e2a ]
In order to avoid all the clients to start reconnecting at the same time
schedule the reconnect dwork with a random jitter of +[0,8] seconds.
Fixes: 6a98d71daea1 ("RDMA/rtrs: client: main functionality")
Link: https://lore.kernel.org/r/20200724111508.15734-2-haris.iqbal@cloud.ionos.com
Signed-off-by: Danil Kipnis <danil.kipnis@cloud.ionos.com>
Signed-off-by: Md Haris Iqbal <haris.iqbal@cloud.ionos.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 47fda651d5af2506deac57d54887cf55ce26e244 ]
The error codes in _ib_modify_qp() are supposed to be negative errno.
Fixes: 7a5c938b9ed0 ("IB/core: Check for rdma_protocol_ib only after validating port_num")
Link: https://lore.kernel.org/r/1595645787-20375-1-git-send-email-liheng40@huawei.com
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Li Heng <liheng40@huawei.com>
Reviewed-by: Parav Pandit <parav@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 5f0b2a6093a4d9aab093964c65083fe801ef1e58 ]
rxe_post_send_kernel() iterates over linked list of wr's, until the
wr->next ptr is NULL. However if we've got an interrupt after last wr is
posted, control may be returned to the code after send completion callback
is executed and wr memory is freed.
As a result, wr->next pointer may contain incorrect value leading to
panic. Store the wr->next on the stack before posting it.
Fixes: 8700e3e7c485 ("Soft RoCE driver")
Link: https://lore.kernel.org/r/20200716190340.23453-1-m.malygin@yadro.com
Signed-off-by: Mikhail Malygin <m.malygin@yadro.com>
Signed-off-by: Sergey Kojushev <s.kojushev@yadro.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit eb7f84e379daad69b4c92538baeaf93bbf493c14 ]
User space should receive the maximum edpm size from kernel driver,
similar to other edpm/ldpm related limits. Add an additional parameter to
the alloc_ucontext_resp structure for the edpm maximum size.
In addition, pass an indication from user-space to kernel
(and not just kernel to user) that the DPM sizes are supported.
This is for supporting backward-forward compatibility between driver and
lib for everything related to DPM transaction and limit sizes.
This should have been part of commit mentioned in Fixes tag.
Link: https://lore.kernel.org/r/20200707063100.3811-3-michal.kalderon@marvell.com
Fixes: 93a3d05f9d68 ("RDMA/qedr: Add kernel capability flags for dpm enabled mode")
Signed-off-by: Ariel Elior <ariel.elior@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit bbe4f4245271bd0f21bf826996c0c5d87a3529c9 ]
In older FW versions the completion flag was treated as the ack flag in
edpm messages. commit ff937b916eb6 ("qed: Add EDPM mode type for user-fw
compatibility") exposed the FW option of setting which mode the QP is in
by adding a flag to the qedr <-> qed API.
This patch adds the qedr <-> libqedr interface so that the libqedr can set
the flag appropriately and qedr can pass it down to FW. Flag is added for
backward compatibility with libqedr.
For older libs, this flag didn't exist and therefore set to zero.
Fixes: ac1b36e55a51 ("qedr: Add support for user context verbs")
Link: https://lore.kernel.org/r/20200707063100.3811-2-michal.kalderon@marvell.com
Signed-off-by: Yuval Bason <yuval.bason@marvell.com>
Signed-off-by: Michal Kalderon <michal.kalderon@marvell.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit acca72e2b031b9fbb4184511072bd246a0abcebc ]
QP's with the same SRQ, working on different CQs and running in parallel
on different CPUs could lead to a race when maintaining the SRQ consumer
count, and leads to FW running out of SRQs. Update the consumer
atomically. Make sure the wqe_prod is updated after the sge_prod due to
FW requirements.
Fixes: 3491c9e799fb ("qedr: Add support for kernel mode SRQ's")
Link: https://lore.kernel.org/r/20200708195526.31040-1-ybason@marvell.com
Signed-off-by: Michal Kalderon <mkalderon@marvell.com>
Signed-off-by: Yuval Basson <ybason@marvell.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 5c99274be8864519328aa74bc550ba410095bc1c ]
In the loopback tests, the following call trace occurs.
Call Trace:
__rxe_do_task+0x1a/0x30 [rdma_rxe]
rxe_qp_destroy+0x61/0xa0 [rdma_rxe]
rxe_destroy_qp+0x20/0x60 [rdma_rxe]
ib_destroy_qp_user+0xcc/0x220 [ib_core]
uverbs_free_qp+0x3c/0xc0 [ib_uverbs]
destroy_hw_idr_uobject+0x24/0x70 [ib_uverbs]
uverbs_destroy_uobject+0x43/0x1b0 [ib_uverbs]
uobj_destroy+0x41/0x70 [ib_uverbs]
__uobj_get_destroy+0x39/0x70 [ib_uverbs]
ib_uverbs_destroy_qp+0x88/0xc0 [ib_uverbs]
ib_uverbs_handler_UVERBS_METHOD_INVOKE_WRITE+0xb9/0xf0 [ib_uverbs]
ib_uverbs_cmd_verbs+0xb16/0xc30 [ib_uverbs]
The root cause is that the actual RDMA connection is not created in the
loopback tests and the rxe_match_dgid will fail randomly.
To fix this call trace which appear in the loopback tests, skip check of
the dgid.
Fixes: 8700e3e7c485 ("Soft RoCE driver")
Link: https://lore.kernel.org/r/20200630123605.446959-1-leon@kernel.org
Signed-off-by: Zhu Yanjun <yanjunz@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
[ Upstream commit 0cb42c0265837fafa2b4f302c8a7fed2631d7869 ]
ib_unregister_device_queued() can only be used by drivers using the new
dealloc_device callback flow, and it has a safety WARN_ON to ensure
drivers are using it properly.
However, if unregister and register are raced there is a special
destruction path that maintains the uniform error handling semantic of
'caller does ib_dealloc_device() on failure'. This requires disabling the
dealloc_device callback which triggers the WARN_ON.
Instead of using NULL to disable the callback use a special function
pointer so the WARN_ON does not trigger.
Fixes: d0899892edd0 ("RDMA/device: Provide APIs from the core code to help unregistration")
Link: https://lore.kernel.org/r/0-v1-a36d512e0a99+762-syz_dealloc_driver_jgg@nvidia.com
Reported-by: syzbot+4088ed905e4ae2b0e13b@syzkaller.appspotmail.com
Suggested-by: Hillf Danton <hdanton@sina.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
|
|
The memory allocated for the DIM wasn't freed in in error unwind path, fix
it by calling to rdma_dim_destroy().
Fixes: da6629793aa6 ("RDMA/core: Provide RDMA DIM support for ULPs")
Link: https://lore.kernel.org/r/20200730082719.1582397-4-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Reviewed-by: Max Gurtovoy <maxg@mellanox.com <mailto:maxg@mellanox.com>>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
HW destroy operation should be last operation after all possible CQ users
completed their work, so move DIM work cancellation before such destroy
call.
Fixes: da6629793aa6 ("RDMA/core: Provide RDMA DIM support for ULPs")
Link: https://lore.kernel.org/r/20200730082719.1582397-3-leon@kernel.org
Reviewed-by: Max Gurtovoy <maxg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
In DCT and RSS RAW QP creation flows, the QP mutex wasn't initialized and
the magic field inside lock was missing. This caused to the following
kernel warning for kernels build with CONFIG_DEBUG_MUTEXES.
DEBUG_LOCKS_WARN_ON(lock->magic != lock)
WARNING: CPU: 3 PID: 16261 at kernel/locking/mutex.c:938 __mutex_lock+0x60e/0x940
Modules linked in: bonding nf_tables ipip tunnel4 geneve ip6_udp_tunnel udp_tunnel ip6_gre ip6_tunnel tunnel6 ip_gre gre ip_tunnel mlx5_ib mlx5_core mlxfw ptp pps_core rdma_ucm ib_uverbs ib_ipoib ib_umad openvswitch nsh xt_MASQUERADE nf_conntrack_netlink nfnetlink iptable_nat xt_addrtype xt_conntrack nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 br_netfilter overlay ib_srp scsi_transport_srp rpcrdma ib_iser libiscsi scsi_transport_iscsi rdma_cm iw_cm ib_cm ib_core [last unloaded: mlxfw]
CPU: 3 PID: 16261 Comm: ib_send_bw Not tainted 5.8.0-rc4_for_upstream_min_debug_2020_07_08_22_04 #1
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014
RIP: 0010:__mutex_lock+0x60e/0x940
Code: c0 0f 84 6d fa ff ff 44 8b 15 4e 9d ba 00 45 85 d2 0f 85 5d fa ff ff 48 c7 c6 f2 de 2b 82 48 c7 c7 f1 8a 2b 82 e8 d2 4d 72 ff <0f> 0b 4c 8b 4d 88 e9 3f fa ff ff f6 c2 04 0f 84 37 fe ff ff 48 89
RSP: 0018:ffff88810bb8b870 EFLAGS: 00010286
RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000000
RDX: ffff88829f1dd880 RSI: 0000000000000000 RDI: ffffffff81192afa
RBP: ffff88810bb8b910 R08: 0000000000000000 R09: 0000000000000028
R10: 0000000000000000 R11: 0000000000003f85 R12: 0000000000000002
R13: ffff88827d8d3ce0 R14: ffffffffa059f615 R15: ffff8882a4d02610
FS: 00007f3f6988e740(0000) GS:ffff8882f5b80000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000556556158000 CR3: 000000010a63c005 CR4: 0000000000360ea0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
? cmd_exec+0x947/0xe60 [mlx5_core]
? __mutex_lock+0x76/0x940
? mlx5_ib_qp_set_counter+0x25/0xa0 [mlx5_ib]
mlx5_ib_qp_set_counter+0x25/0xa0 [mlx5_ib]
mlx5_ib_counter_bind_qp+0x9b/0xe0 [mlx5_ib]
__rdma_counter_bind_qp+0x6b/0xa0 [ib_core]
rdma_counter_bind_qp_auto+0x363/0x520 [ib_core]
_ib_modify_qp+0x316/0x580 [ib_core]
ib_modify_qp_with_udata+0x19/0x30 [ib_core]
modify_qp+0x4c4/0x600 [ib_uverbs]
ib_uverbs_ex_modify_qp+0x87/0xe0 [ib_uverbs]
ib_uverbs_handler_UVERBS_METHOD_INVOKE_WRITE+0x129/0x1c0 [ib_uverbs]
ib_uverbs_cmd_verbs.isra.5+0x5d5/0x11f0 [ib_uverbs]
? ib_uverbs_handler_UVERBS_METHOD_QUERY_CONTEXT+0x120/0x120 [ib_uverbs]
? lock_acquire+0xb9/0x3a0
? ib_uverbs_ioctl+0xd0/0x210 [ib_uverbs]
? ib_uverbs_ioctl+0x175/0x210 [ib_uverbs]
ib_uverbs_ioctl+0x14b/0x210 [ib_uverbs]
? ib_uverbs_ioctl+0xd0/0x210 [ib_uverbs]
ksys_ioctl+0x234/0x7d0
? exc_page_fault+0x202/0x640
? do_syscall_64+0x1f/0x2e0
__x64_sys_ioctl+0x16/0x20
do_syscall_64+0x59/0x2e0
? asm_exc_page_fault+0x8/0x30
? rcu_read_lock_sched_held+0x52/0x60
entry_SYSCALL_64_after_hwframe+0x44/0xa9
Fixes: b4aaa1f0b415 ("IB/mlx5: Handle type IB_QPT_DRIVER when creating a QP")
Link: https://lore.kernel.org/r/20200730082719.1582397-2-leon@kernel.org
Reviewed-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
The lookaside count is improperly initialized to the size of the
Receive Queue with the additional +1. In the traces below, the
RQ size is 384, so the count was set to 385.
The lookaside count is then rarely refreshed. Note the high and
incorrect count in the trace below:
rvt_get_rwqe: [hfi1_0] wqe ffffc900078e9008 wr_id 55c7206d75a0 qpn c
qpt 2 pid 3018 num_sge 1 head 1 tail 0, count 385
rvt_get_rwqe: (hfi1_rc_rcv+0x4eb/0x1480 [hfi1] <- rvt_get_rwqe) ret=0x1
The head,tail indicate there is only one RWQE posted although the count
says 385 and we correctly return the element 0.
The next call to rvt_get_rwqe with the decremented count:
rvt_get_rwqe: [hfi1_0] wqe ffffc900078e9058 wr_id 0 qpn c
qpt 2 pid 3018 num_sge 0 head 1 tail 1, count 384
rvt_get_rwqe: (hfi1_rc_rcv+0x4eb/0x1480 [hfi1] <- rvt_get_rwqe) ret=0x1
Note that the RQ is empty (head == tail) yet we return the RWQE at tail 1,
which is not valid because of the bogus high count.
Best case, the RWQE has never been posted and the rc logic sees an RWQE
that is too small (all zeros) and puts the QP into an error state.
In the worst case, a server slow at posting receive buffers might fool
rvt_get_rwqe() into fetching an old RWQE and corrupt memory.
Fix by deleting the faulty initialization code and creating an
inline to fetch the posted count and convert all callers to use
new inline.
Fixes: f592ae3c999f ("IB/rdmavt: Fracture single lock used for posting and processing RWQEs")
Link: https://lore.kernel.org/r/20200728183848.22226.29132.stgit@awfm-01.aw.intel.com
Reported-by: Zhaojuan Guo <zguo@redhat.com>
Cc: <stable@vger.kernel.org> # 5.4.x
Reviewed-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Tested-by: Honggang Li <honli@redhat.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
Scatter CQE feature relies on two flags MLX5_QP_FLAG_SCATTER_CQE and
MLX5_QP_FLAG_ALLOW_SCATTER_CQE, both of them can be provided without
relation to device capability.
Relax global validity check to allow MLX5_QP_FLAG_ALLOW_SCATTER_CQE QP
flag.
Existing user applications are failing on this new validity check.
Fixes: 90ecb37a751b ("RDMA/mlx5: Change scatter CQE flag to be set like other vendor flags")
Fixes: 37518fa49f76 ("RDMA/mlx5: Process all vendor flags in one place")
Link: https://lore.kernel.org/r/20200728120255.805733-1-leon@kernel.org
Reviewed-by: Artemy Kovalyov <artemyko@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
destroy_prefetch_work() must always be called if the work is not going
to be queued. The num_sge also should have been set to i, not i-1
which avoids the condition where it shouldn't have been called in the
first place.
Cc: stable@vger.kernel.org
Fixes: fb985e278a30 ("RDMA/mlx5: Use SRCU properly in ODP prefetch")
Link: https://lore.kernel.org/r/20200727095712.495652-1-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
These are missing throughout ucma, it harmlessly copies garbage from
userspace, but in this new code which uses min to compute the copy length
it can result in uninitialized stack memory. Check for minimum length at
the very start.
BUG: KMSAN: uninit-value in ucma_connect+0x2aa/0xab0 drivers/infiniband/core/ucma.c:1091
CPU: 0 PID: 8457 Comm: syz-executor069 Not tainted 5.8.0-rc5-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
Call Trace:
__dump_stack lib/dump_stack.c:77 [inline]
dump_stack+0x1df/0x240 lib/dump_stack.c:118
kmsan_report+0xf7/0x1e0 mm/kmsan/kmsan_report.c:121
__msan_warning+0x58/0xa0 mm/kmsan/kmsan_instr.c:215
ucma_connect+0x2aa/0xab0 drivers/infiniband/core/ucma.c:1091
ucma_write+0x5c5/0x630 drivers/infiniband/core/ucma.c:1764
do_loop_readv_writev fs/read_write.c:737 [inline]
do_iter_write+0x710/0xdc0 fs/read_write.c:1020
vfs_writev fs/read_write.c:1091 [inline]
do_writev+0x42d/0x8f0 fs/read_write.c:1134
__do_sys_writev fs/read_write.c:1207 [inline]
__se_sys_writev+0x9b/0xb0 fs/read_write.c:1204
__x64_sys_writev+0x4a/0x70 fs/read_write.c:1204
do_syscall_64+0xb0/0x150 arch/x86/entry/common.c:386
entry_SYSCALL_64_after_hwframe+0x44/0xa9
Fixes: 34e2ab57a911 ("RDMA/ucma: Extend ucma_connect to receive ECE parameters")
Fixes: 0cb15372a615 ("RDMA/cma: Connect ECE to rdma_accept")
Link: https://lore.kernel.org/r/0-v1-d5b86dab17dc+28c25-ucma_syz_min_jgg@nvidia.com
Reported-by: syzbot+086ab5ca9eafd2379aa6@syzkaller.appspotmail.com
Reported-by: syzbot+7446526858b83c8828b2@syzkaller.appspotmail.com
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
Prefetch work in mlx5_ib_prefetch_mr_work can be queued and able to run
concurrently with destruction of the implicit MR. The num_deferred_work
was intended to serialize this, but there is a race:
CPU0 CPU1
mlx5_ib_free_implicit_mr()
xa_erase(odp_mkeys)
synchronize_srcu()
__xa_erase(implicit_children)
mlx5_ib_prefetch_mr_work()
pagefault_mr()
pagefault_implicit_mr()
implicit_get_child_mr()
xa_cmpxchg()
atomic_dec_and_test(num_deferred_mr)
wait_event(imr->q_deferred_work)
ib_umem_odp_release(odp_imr)
kfree(odp_imr)
At this point in mlx5_ib_free_implicit_mr() the implicit_children list is
supposed to be empty forever so that destroy_unused_implicit_child_mr()
and related are not and will not be running.
Since it is not empty the destroy_unused_implicit_child_mr() flow ends up
touching deallocated memory as mlx5_ib_free_implicit_mr() already tore down the
imr parent.
The solution is to flush out the prefetch wq by driving num_deferred_work
to zero after creation of new prefetch work is blocked.
Fixes: 5256edcb98a1 ("RDMA/mlx5: Rework implicit ODP destroy")
Link: https://lore.kernel.org/r/20200719065435.130722-1-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
cm.lock must be held while accessing remote_sidr_table. This fixes the
below NULL pointer dereference.
BUG: kernel NULL pointer dereference, address: 0000000000000000
#PF: supervisor write access in kernel mode
#PF: error_code(0x0002) - not-present page
PGD 0 P4D 0
Oops: 0002 [#1] SMP PTI
CPU: 2 PID: 7288 Comm: udaddy Not tainted 5.7.0_for_upstream_perf_2020_06_09_15_14_20_38 #1
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014
RIP: 0010:rb_erase+0x10d/0x360
Code: 00 00 00 48 89 c1 48 89 d0 48 8b 50 08 48 39 ca 74 48 f6 02 01 75 af 48 8b 7a 10 48 89 c1 48 83 c9 01 48 89 78 08 48 89 42 10 <48> 89 0f 48 8b 08 48 89 0a 48 83 e1 fc 48 89 10 0f 84 b1 00 00 00
RSP: 0018:ffffc90000f77c30 EFLAGS: 00010086
RAX: ffff8883df27d458 RBX: ffff8883df27da58 RCX: ffff8883df27d459
RDX: ffff8883d183fa58 RSI: ffffffffa01e8d00 RDI: 0000000000000000
RBP: ffff8883d62ac800 R08: 0000000000000000 R09: 00000000000000ce
R10: 000000000000000a R11: 0000000000000000 R12: ffff8883df27da00
R13: ffffc90000f77c98 R14: 0000000000000130 R15: 0000000000000000
FS: 00007f009f877740(0000) GS:ffff8883f1a00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000000 CR3: 00000003d467e003 CR4: 0000000000160ee0
Call Trace:
cm_send_sidr_rep_locked+0x15a/0x1a0 [ib_cm]
ib_send_cm_sidr_rep+0x2b/0x50 [ib_cm]
cma_send_sidr_rep+0x8b/0xe0 [rdma_cm]
__rdma_accept+0x21d/0x2b0 [rdma_cm]
? ucma_get_ctx+0x2b/0xe0 [rdma_ucm]
? _copy_from_user+0x30/0x60
ucma_accept+0x13e/0x1e0 [rdma_ucm]
ucma_write+0xb4/0x130 [rdma_ucm]
vfs_write+0xad/0x1a0
ksys_write+0x9d/0xb0
do_syscall_64+0x48/0x130
entry_SYSCALL_64_after_hwframe+0x44/0xa9
RIP: 0033:0x7f009ef60924
Code: 00 f7 d8 64 89 02 48 c7 c0 ff ff ff ff eb b7 0f 1f 80 00 00 00 00 8b 05 2a ef 2c 00 48 63 ff 85 c0 75 13 b8 01 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 54 f3 c3 66 90 55 53 48 89 d5 48 89 f3 48 83
RSP: 002b:00007fff843edf38 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
RAX: ffffffffffffffda RBX: 000055743042e1d0 RCX: 00007f009ef60924
RDX: 0000000000000130 RSI: 00007fff843edf40 RDI: 0000000000000003
RBP: 00007fff843ee0e0 R08: 0000000000000000 R09: 0000557430433090
R10: 0000000000000001 R11: 0000000000000246 R12: 0000000000000000
R13: 00007fff843edf40 R14: 000000000000038c R15: 00000000ffffff00
CR2: 0000000000000000
Fixes: 6a8824a74bc9 ("RDMA/cm: Allow ib_send_cm_sidr_rep() to be done under lock")
Link: https://lore.kernel.org/r/20200716105519.1424266-1-leon@kernel.org
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
The FD should not be installed until all of the setup is completed as the
fd_install() transfers ownership of the kref to the FD table. A thread can
race a close() and trigger concurrent rdma_alloc_commit_uobject() and
uverbs_uobject_fd_release() which, at least, triggers a safety WARN_ON:
WARNING: CPU: 4 PID: 6913 at drivers/infiniband/core/rdma_core.c:768 uverbs_uobject_fd_release+0x202/0x230
Kernel panic - not syncing: panic_on_warn set ...
CPU: 4 PID: 6913 Comm: syz-executor.3 Not tainted 5.7.0-rc2 #22
Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014
[..]
RIP: 0010:uverbs_uobject_fd_release+0x202/0x230
Code: fe 4c 89 e7 e8 af 23 fe ff e9 2a ff ff ff e8 c5 fa 61 fe be 03 00 00 00 4c 89 e7 e8 68 eb f5 fe e9 13 ff ff ff e8 ae fa 61 fe <0f> 0b eb ac e8 e5 aa 3c fe e8 50 2b 86 fe e9 6a fe ff ff e8 46 2b
RSP: 0018:ffffc90008117d88 EFLAGS: 00010293
RAX: ffff88810e146580 RBX: 1ffff92001022fb1 RCX: ffffffff82d5b902
RDX: 0000000000000000 RSI: 0000000000000004 RDI: ffff88811951b040
RBP: ffff88811951b000 R08: ffffed10232a3609 R09: ffffed10232a3609
R10: ffff88811951b043 R11: 0000000000000001 R12: ffff888100a7c600
R13: ffff888100a7c650 R14: ffffc90008117da8 R15: ffffffff82d5b700
? __uverbs_cleanup_ufile+0x270/0x270
? uverbs_uobject_fd_release+0x202/0x230
? uverbs_uobject_fd_release+0x202/0x230
? __uverbs_cleanup_ufile+0x270/0x270
? locks_remove_file+0x282/0x3d0
? security_file_free+0xaa/0xd0
__fput+0x2be/0x770
task_work_run+0x10e/0x1b0
exit_to_usermode_loop+0x145/0x170
do_syscall_64+0x2d0/0x390
? prepare_exit_to_usermode+0x17a/0x230
entry_SYSCALL_64_after_hwframe+0x44/0xa9
RIP: 0033:0x414da7
Code: 00 00 0f 05 48 3d 00 f0 ff ff 77 3f f3 c3 0f 1f 44 00 00 53 89 fb 48 83 ec 10 e8 f4 fb ff ff 89 df 89 c2 b8 03 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 2b 89 d7 89 44 24 0c e8 36 fc ff ff 8b 44 24
RSP: 002b:00007fff39d379d0 EFLAGS: 00000293 ORIG_RAX: 0000000000000003
RAX: 0000000000000000 RBX: 0000000000000003 RCX: 0000000000414da7
RDX: 0000000000000000 RSI: 0000000000000001 RDI: 0000000000000003
RBP: 00007fff39d37a3c R08: 0000000400000000 R09: 0000000400000000
R10: 00007fff39d37910 R11: 0000000000000293 R12: 0000000000000001
R13: 0000000000000001 R14: 0000000000000000 R15: 0000000000000003
Reorder so that fd_install() is the last thing done in
rdma_alloc_commit_uobject().
Fixes: aba94548c9e4 ("IB/uverbs: Move the FD uobj type struct file allocation to alloc_commit")
Link: https://lore.kernel.org/r/20200716102059.1420681-1-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
ROCE uses "VA % buf_page_size" to caclulate the offset in the PBL's first
page, the actual PA corresponding to the MR's VA is equal to MR's PA plus
this offset. The first PA in PBL has already been aligned to PAGE_SIZE
after calling ib_umem_get(), but the MR's VA may not. If the buf_page_size
is smaller than the PAGE_SIZE, this will lead the HW to access the wrong
memory because the offset is smaller than expected.
Fixes: 9b2cf76c9f05 ("RDMA/hns: Optimize PBL buffer allocation process")
Link: https://lore.kernel.org/r/1594726935-45666-1-git-send-email-liweihang@huawei.com
Signed-off-by: Xi Wang <wangxi11@huawei.com>
Signed-off-by: Weihang Li <liweihang@huawei.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
The RoCE Engine will schedule to another QP after one has sent
(2 ^ lp_pktn_ini) packets. lp_pktn_ini is set in QPC and should be
calculated from 2 factors:
1. current MTU as a integer
2. the RoCE Engine's maximum slice length 64KB
But the driver use MTU as a enum ib_mtu and the max inline capability, the
lp_pktn_ini will be much bigger than expected which may cause traffic of
some QPs to never get scheduled.
Fixes: b713128de7a1 ("RDMA/hns: Adjust lp_pktn_ini dynamically")
Link: https://lore.kernel.org/r/1594726138-49294-1-git-send-email-liweihang@huawei.com
Signed-off-by: Weihang Li <liweihang@huawei.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
SRQ table is accessed both from interrupt and process context,
therefore we must use xa_lock_irq.
inconsistent {IN-HARDIRQ-W} -> {HARDIRQ-ON-W} usage.
kworker/u17:9/8573 takes:
ffff8883e3503d30 (&xa->xa_lock#13){?...}-{2:2}, at: mlx5_cmd_get_srq+0x18/0x70 [mlx5_ib]
{IN-HARDIRQ-W} state was registered at:
lock_acquire+0xb9/0x3a0
_raw_spin_lock+0x25/0x30
srq_event_notifier+0x2b/0xc0 [mlx5_ib]
notifier_call_chain+0x45/0x70
__atomic_notifier_call_chain+0x69/0x100
forward_event+0x36/0xc0 [mlx5_core]
notifier_call_chain+0x45/0x70
__atomic_notifier_call_chain+0x69/0x100
mlx5_eq_async_int+0xc5/0x160 [mlx5_core]
notifier_call_chain+0x45/0x70
__atomic_notifier_call_chain+0x69/0x100
mlx5_irq_int_handler+0x19/0x30 [mlx5_core]
__handle_irq_event_percpu+0x43/0x2a0
handle_irq_event_percpu+0x30/0x70
handle_irq_event+0x34/0x60
handle_edge_irq+0x7c/0x1b0
do_IRQ+0x60/0x110
ret_from_intr+0x0/0x2a
default_idle+0x34/0x160
do_idle+0x1ec/0x220
cpu_startup_entry+0x19/0x20
start_secondary+0x153/0x1a0
secondary_startup_64+0xa4/0xb0
irq event stamp: 20907
hardirqs last enabled at (20907): _raw_spin_unlock_irq+0x24/0x30
hardirqs last disabled at (20906): _raw_spin_lock_irq+0xf/0x40
softirqs last enabled at (20746): __do_softirq+0x2c9/0x436
softirqs last disabled at (20681): irq_exit+0xb3/0xc0
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0
----
lock(&xa->xa_lock#13);
<Interrupt>
lock(&xa->xa_lock#13);
*** DEADLOCK ***
2 locks held by kworker/u17:9/8573:
#0: ffff888295218d38 ((wq_completion)mlx5_ib_page_fault){+.+.}-{0:0}, at: process_one_work+0x1f1/0x5f0
#1: ffff888401647e78 ((work_completion)(&pfault->work)){+.+.}-{0:0}, at: process_one_work+0x1f1/0x5f0
stack backtrace:
CPU: 0 PID: 8573 Comm: kworker/u17:9 Tainted: GO 5.7.0_for_upstream_min_debug_2020_06_14_11_31_46_41 #1
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014
Workqueue: mlx5_ib_page_fault mlx5_ib_eqe_pf_action [mlx5_ib]
Call Trace:
dump_stack+0x71/0x9b
mark_lock+0x4f2/0x590
? print_shortest_lock_dependencies+0x200/0x200
__lock_acquire+0xa00/0x1eb0
lock_acquire+0xb9/0x3a0
? mlx5_cmd_get_srq+0x18/0x70 [mlx5_ib]
_raw_spin_lock+0x25/0x30
? mlx5_cmd_get_srq+0x18/0x70 [mlx5_ib]
mlx5_cmd_get_srq+0x18/0x70 [mlx5_ib]
mlx5_ib_eqe_pf_action+0x257/0xa30 [mlx5_ib]
? process_one_work+0x209/0x5f0
process_one_work+0x27b/0x5f0
? __schedule+0x280/0x7e0
worker_thread+0x2d/0x3c0
? process_one_work+0x5f0/0x5f0
kthread+0x111/0x130
? kthread_park+0x90/0x90
ret_from_fork+0x24/0x30
Fixes: e126ba97dba9 ("mlx5: Add driver for Mellanox Connect-IB adapters")
Link: https://lore.kernel.org/r/20200712102641.15210-1-leon@kernel.org
Signed-off-by: Maor Gottlieb <maorg@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
ib_pd is accessed internally during destroy of the TIR/TIS, but PD
can be not set yet. This leading to the following kernel panic.
BUG: kernel NULL pointer dereference, address: 0000000000000074
PGD 8000000079eaa067 P4D 8000000079eaa067 PUD 7ae81067 PMD 0 Oops: 0000 [#1] SMP PTI
CPU: 1 PID: 709 Comm: syz-executor.0 Not tainted 5.8.0-rc3 #41 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.12.1-0-ga5cab58e9a3f-prebuilt.qemu.org 04/01/2014
RIP: 0010:destroy_raw_packet_qp_tis drivers/infiniband/hw/mlx5/qp.c:1189 [inline]
RIP: 0010:destroy_raw_packet_qp drivers/infiniband/hw/mlx5/qp.c:1527 [inline]
RIP: 0010:destroy_qp_common+0x2ca/0x4f0 drivers/infiniband/hw/mlx5/qp.c:2397
Code: 00 85 c0 74 2e e8 56 18 55 ff 48 8d b3 28 01 00 00 48 89 ef e8 d7 d3 ff ff 48 8b 43 08 8b b3 c0 01 00 00 48 8b bd a8 0a 00 00 <0f> b7 50 74 e8 0d 6a fe ff e8 28 18 55 ff 49 8d 55 50 4c 89 f1 48
RSP: 0018:ffffc900007bbac8 EFLAGS: 00010293
RAX: 0000000000000000 RBX: ffff88807949e800 RCX: 0000000000000998
RDX: 0000000000000000 RSI: 0000000000000008 RDI: ffff88807c180140
RBP: ffff88807b50c000 R08: 000000000002d379 R09: ffffc900007bba00
R10: 0000000000000001 R11: 000000000002d358 R12: ffff888076f37000
R13: ffff88807949e9c8 R14: ffffc900007bbe08 R15: ffff888076f37000
FS: 00000000019bf940(0000) GS:ffff88807dd00000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000074 CR3: 0000000076d68004 CR4: 0000000000360ee0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
mlx5_ib_create_qp+0xf36/0xf90 drivers/infiniband/hw/mlx5/qp.c:3014
_ib_create_qp drivers/infiniband/core/core_priv.h:333 [inline]
create_qp+0x57f/0xd20 drivers/infiniband/core/uverbs_cmd.c:1443
ib_uverbs_create_qp+0xcf/0x100 drivers/infiniband/core/uverbs_cmd.c:1564
ib_uverbs_write+0x5fa/0x780 drivers/infiniband/core/uverbs_main.c:664
__vfs_write+0x3f/0x90 fs/read_write.c:495
vfs_write+0xc7/0x1f0 fs/read_write.c:559
ksys_write+0x5e/0x110 fs/read_write.c:612
do_syscall_64+0x3e/0x70 arch/x86/entry/common.c:359
entry_SYSCALL_64_after_hwframe+0x44/0xa9
RIP: 0033:0x466479
Code: Bad RIP value.
RSP: 002b:00007ffd057b62b8 EFLAGS: 00000246 ORIG_RAX: 0000000000000001
RAX: ffffffffffffffda RBX: 000000000073bf00 RCX: 0000000000466479
RDX: 0000000000000070 RSI: 0000000020000240 RDI: 0000000000000003
RBP: 00000000019bf8fc R08: 0000000000000000 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000246 R12: 00000000ffffffff
R13: 0000000000000bf6 R14: 00000000004cb859 R15: 00000000006fefc0
Fixes: 6c41965d647a ("RDMA/mlx5: Don't access ib_qp fields in internal destroy QP path")
Link: https://lore.kernel.org/r/20200707110612.882962-4-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
Some released FW versions mistakenly don't set the capability that 50G per
lane link-modes are supported for VFs (ptys_extended_ethernet capability
bit).
Use PTYS.ext_eth_proto_capability instead, as this indication is always
accurate. If PTYS.ext_eth_proto_capability is valid
(has a non-zero value) conclude that the HCA supports 50G per lane.
Otherwise, conclude that the HCA doesn't support 50G per lane.
Fixes: 08e8676f1607 ("IB/mlx5: Add support for 50Gbps per lane link modes")
Link: https://lore.kernel.org/r/20200707110612.882962-3-leon@kernel.org
Signed-off-by: Aya Levin <ayal@mellanox.com>
Reviewed-by: Eran Ben Elisha <eranbe@mellanox.com>
Reviewed-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
Move the initialization of the vendor_part_id to be before calling
ib_register_device(), this is needed because the query_device() callback
is called from the context of ib_register_device() before initializing the
vendor_part_id, so the reported value is wrong.
Fixes: bdcf26bf9b3a ("rdma/siw: network and RDMA core interface")
Link: https://lore.kernel.org/r/20200707130931.444724-1-kamalheib1@gmail.com
Signed-off-by: Kamal Heib <kamalheib1@gmail.com>
Reviewed-by: Bernard Metzler <bmt@zurich.ibm.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
There is a race condition where ib_nl_make_request() inserts the request
data into the linked list but the timer in ib_nl_request_timeout() can see
it and destroy it before ib_nl_send_msg() is done touching it. This could
happen, for instance, if there is a long delay allocating memory during
nlmsg_new()
This causes a use-after-free in the send_mad() thread:
[<ffffffffa02f43cb>] ? ib_pack+0x17b/0x240 [ib_core]
[ <ffffffffa032aef1>] ib_sa_path_rec_get+0x181/0x200 [ib_sa]
[<ffffffffa0379db0>] rdma_resolve_route+0x3c0/0x8d0 [rdma_cm]
[<ffffffffa0374450>] ? cma_bind_port+0xa0/0xa0 [rdma_cm]
[<ffffffffa040f850>] ? rds_rdma_cm_event_handler_cmn+0x850/0x850 [rds_rdma]
[<ffffffffa040f22c>] rds_rdma_cm_event_handler_cmn+0x22c/0x850 [rds_rdma]
[<ffffffffa040f860>] rds_rdma_cm_event_handler+0x10/0x20 [rds_rdma]
[<ffffffffa037778e>] addr_handler+0x9e/0x140 [rdma_cm]
[<ffffffffa026cdb4>] process_req+0x134/0x190 [ib_addr]
[<ffffffff810a02f9>] process_one_work+0x169/0x4a0
[<ffffffff810a0b2b>] worker_thread+0x5b/0x560
[<ffffffff810a0ad0>] ? flush_delayed_work+0x50/0x50
[<ffffffff810a68fb>] kthread+0xcb/0xf0
[<ffffffff816ec49a>] ? __schedule+0x24a/0x810
[<ffffffff816ec49a>] ? __schedule+0x24a/0x810
[<ffffffff810a6830>] ? kthread_create_on_node+0x180/0x180
[<ffffffff816f25a7>] ret_from_fork+0x47/0x90
[<ffffffff810a6830>] ? kthread_create_on_node+0x180/0x180
The ownership rule is once the request is on the list, ownership transfers
to the list and the local thread can't touch it any more, just like for
the normal MAD case in send_mad().
Thus, instead of adding before send and then trying to delete after on
errors, move the entire thing under the spinlock so that the send and
update of the lists are atomic to the conurrent threads. Lightly reoganize
things so spinlock safe memory allocations are done in the final NL send
path and the rest of the setup work is done before and outside the lock.
Fixes: 3ebd2fd0d011 ("IB/sa: Put netlink request into the request list before sending")
Link: https://lore.kernel.org/r/1592964789-14533-1-git-send-email-divya.indi@oracle.com
Signed-off-by: Divya Indi <divya.indi@oracle.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
The workqueue link_wq should only be destroyed when the hfi1 driver is
unloaded, not when the device is shut down.
Fixes: 71d47008ca1b ("IB/hfi1: Create workqueue for link events")
Link: https://lore.kernel.org/r/20200623204053.107638.70315.stgit@awfm-01.aw.intel.com
Cc: <stable@vger.kernel.org>
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
The workqueue hfi1_wq is destroyed in function shutdown_device(), which is
called by either shutdown_one() or remove_one(). The function
shutdown_one() is called when the kernel is rebooted while remove_one() is
called when the hfi1 driver is unloaded. When the kernel is rebooted,
hfi1_wq is destroyed while all qps are still active, leading to a kernel
crash:
BUG: unable to handle kernel NULL pointer dereference at 0000000000000102
IP: [<ffffffff94cb7b02>] __queue_work+0x32/0x3e0
PGD 0
Oops: 0000 [#1] SMP
Modules linked in: dm_round_robin nvme_rdma(OE) nvme_fabrics(OE) nvme_core(OE) ib_isert iscsi_target_mod target_core_mod ib_ucm mlx4_ib iTCO_wdt iTCO_vendor_support mxm_wmi sb_edac intel_powerclamp coretemp intel_rapl iosf_mbi kvm rpcrdma sunrpc irqbypass crc32_pclmul ghash_clmulni_intel rdma_ucm aesni_intel ib_uverbs lrw gf128mul opa_vnic glue_helper ablk_helper ib_iser cryptd ib_umad rdma_cm iw_cm ses enclosure libiscsi scsi_transport_sas pcspkr joydev ib_ipoib(OE) scsi_transport_iscsi ib_cm sg ipmi_ssif mei_me lpc_ich i2c_i801 mei ioatdma ipmi_si dm_multipath ipmi_devintf ipmi_msghandler wmi acpi_pad acpi_power_meter hangcheck_timer ip_tables ext4 mbcache jbd2 mlx4_en sd_mod crc_t10dif crct10dif_generic mgag200 drm_kms_helper syscopyarea sysfillrect sysimgblt fb_sys_fops ttm hfi1(OE)
crct10dif_pclmul crct10dif_common crc32c_intel drm ahci mlx4_core libahci rdmavt(OE) igb megaraid_sas ib_core libata drm_panel_orientation_quirks ptp pps_core devlink dca i2c_algo_bit dm_mirror dm_region_hash dm_log dm_mod
CPU: 19 PID: 0 Comm: swapper/19 Kdump: loaded Tainted: G OE ------------ 3.10.0-957.el7.x86_64 #1
Hardware name: Phegda X2226A/S2600CW, BIOS SE5C610.86B.01.01.0024.021320181901 02/13/2018
task: ffff8a799ba0d140 ti: ffff8a799bad8000 task.ti: ffff8a799bad8000
RIP: 0010:[<ffffffff94cb7b02>] [<ffffffff94cb7b02>] __queue_work+0x32/0x3e0
RSP: 0018:ffff8a90dde43d80 EFLAGS: 00010046
RAX: 0000000000000082 RBX: 0000000000000086 RCX: 0000000000000000
RDX: ffff8a90b924fcb8 RSI: 0000000000000000 RDI: 000000000000001b
RBP: ffff8a90dde43db8 R08: ffff8a799ba0d6d8 R09: ffff8a90dde53900
R10: 0000000000000002 R11: ffff8a90dde43de8 R12: ffff8a90b924fcb8
R13: 000000000000001b R14: 0000000000000000 R15: ffff8a90d2890000
FS: 0000000000000000(0000) GS:ffff8a90dde40000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 0000000000000102 CR3: 0000001a70410000 CR4: 00000000001607e0
Call Trace:
[<ffffffff94cb8105>] queue_work_on+0x45/0x50
[<ffffffffc03f781e>] _hfi1_schedule_send+0x6e/0xc0 [hfi1]
[<ffffffffc03f78a2>] hfi1_schedule_send+0x32/0x70 [hfi1]
[<ffffffffc02cf2d9>] rvt_rc_timeout+0xe9/0x130 [rdmavt]
[<ffffffff94ce563a>] ? trigger_load_balance+0x6a/0x280
[<ffffffffc02cf1f0>] ? rvt_free_qpn+0x40/0x40 [rdmavt]
[<ffffffff94ca7f58>] call_timer_fn+0x38/0x110
[<ffffffffc02cf1f0>] ? rvt_free_qpn+0x40/0x40 [rdmavt]
[<ffffffff94caa3bd>] run_timer_softirq+0x24d/0x300
[<ffffffff94ca0f05>] __do_softirq+0xf5/0x280
[<ffffffff9537832c>] call_softirq+0x1c/0x30
[<ffffffff94c2e675>] do_softirq+0x65/0xa0
[<ffffffff94ca1285>] irq_exit+0x105/0x110
[<ffffffff953796c8>] smp_apic_timer_interrupt+0x48/0x60
[<ffffffff95375df2>] apic_timer_interrupt+0x162/0x170
<EOI>
[<ffffffff951adfb7>] ? cpuidle_enter_state+0x57/0xd0
[<ffffffff951ae10e>] cpuidle_idle_call+0xde/0x230
[<ffffffff94c366de>] arch_cpu_idle+0xe/0xc0
[<ffffffff94cfc3ba>] cpu_startup_entry+0x14a/0x1e0
[<ffffffff94c57db7>] start_secondary+0x1f7/0x270
[<ffffffff94c000d5>] start_cpu+0x5/0x14
The solution is to destroy the workqueue only when the hfi1 driver is
unloaded, not when the device is shut down. In addition, when the device
is shut down, no more work should be scheduled on the workqueues and the
workqueues are flushed.
Fixes: 8d3e71136a08 ("IB/{hfi1, qib}: Add handling of kernel restart")
Link: https://lore.kernel.org/r/20200623204047.107638.77646.stgit@awfm-01.aw.intel.com
Cc: <stable@vger.kernel.org>
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
Legacy IPoIB sets IB_QP_CREATE_NETIF_QP QP create flag and because mlx5
doesn't use this flag, the process_create_flags() failed to create IPoIB
QPs.
Fixes: 2978975ce7f1 ("RDMA/mlx5: Process create QP flags in one place")
Link: https://lore.kernel.org/r/20200630122147.445847-1-leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
Clang warns:
drivers/infiniband/hw/hfi1/qp.c:198:9: warning: implicit conversion from enumeration type 'enum opa_mtu' to different enumeration type 'enum ib_mtu' [-Wenum-conversion]
mtu = OPA_MTU_8192;
~ ^~~~~~~~~~~~
enum opa_mtu extends enum ib_mtu. There are typically two ways to deal
with this:
* Remove the expected types and just use 'int' for all parameters and
types.
* Explicitly cast the enums between each other.
This driver chooses to do the later so do the same thing here.
Fixes: 6d72344cf6c4 ("IB/ipoib: Increase ipoib Datagram mode MTU's upper limit")
Link: https://lore.kernel.org/r/20200623005224.492239-1-natechancellor@gmail.com
Link: https://github.com/ClangBuiltLinux/linux/issues/1062
Link: https://lore.kernel.org/linux-rdma/20200527040350.GA3118979@ubuntu-s3-xlarge-x86/
Signed-off-by: Nathan Chancellor <natechancellor@gmail.com>
Acked-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
When running iperf in a two host configuration the following trace can
occur:
[ 319.728730] NETDEV WATCHDOG: ib0 (hfi1): transmit queue 0 timed out
The issue happens because the current implementation relies on the netif
txq being stopped to control the flushing of the tx list.
There are two resources that the transmit logic can wait on and stop the
txq:
- SDMA descriptors
- Ring space to hold completions
The ring space is tested on the sending side and relieved when the ring is
consumed in the napi tx reaping.
Unfortunately, that reaping can run conncurrently with the workqueue
flushing of the txlist. If the txq is started just before the workitem
executes, the txlist will never be flushed, leading to the txq being
stuck.
Fix by:
- Adding sleep/wakeup wrappers
* Use an atomic to control the call to the netif routines inside the
wrappers
- Use another atomic to record ring space exhaustion
* Only wakeup when the a ring space exhaustion has happened and it
relieved
Add additional wrappers to clarify the ring space resource handling.
Fixes: d99dc602e2a5 ("IB/hfi1: Add functions to transmit datagram ipoib packets")
Link: https://lore.kernel.org/r/20200623204327.108092.4024.stgit@awfm-01.aw.intel.com
Reviewed-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
The current code mishandles -EBUSY in two ways:
- The flow change doesn't test the return from the flush and runs on to
process the current packet racing with the wakeup processing
- The -EBUSY handling for a single packet inserts the tx into the txlist
after the submit call, racing with the same wakeup processing
Fix the first by dropping the skb and returning NETDEV_TX_OK.
Fix the second by insuring the the list entry within the txreq is inited
when allocated. This enables the sleep routine to detect that the txreq
has used the non-list api and queue the packet to the txlist.
Both flaws can lead to having the flushing thread executing in causing two
threads to manipulate the txlist.
Fixes: d99dc602e2a5 ("IB/hfi1: Add functions to transmit datagram ipoib packets")
Link: https://lore.kernel.org/r/20200623204321.108092.83898.stgit@awfm-01.aw.intel.com
Reviewed-by: Kaike Wan <kaike.wan@intel.com>
Signed-off-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|
|
When the try_module_get calls were removed from opening and closing of the
i2c debugfs file, the corresponding module_put calls were missed. This
results in an inaccurate module use count that requires a power cycle to
fix.
Fixes: 09fbca8e6240 ("IB/hfi1: No need to use try_module_get for debugfs")
Link: https://lore.kernel.org/r/20200623203230.106975.76240.stgit@awfm-01.aw.intel.com
Cc: <stable@vger.kernel.org>
Reviewed-by: Kaike Wan <kaike.wan@intel.com>
Reviewed-by: Mike Marciniszyn <mike.marciniszyn@intel.com>
Signed-off-by: Dennis Dalessandro <dennis.dalessandro@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
|