linux-stable.git/drivers/infiniband/ulp, branch v5.4.151 Linux kernel stable tree RDMA/ipoib: Fix warning caused by destroying non-initial netns 2021-06-16T09:59:44+00:00 Kamal Heib kamalheib1@gmail.com 2021-05-25T15:01:34+00:00 64f1fb6acc2ab95982fc4334f351d7576c26f313 commit a3e74fb9247cd530dca246699d5eb5a691884d32 upstream. After the commit 5ce2dced8e95 ("RDMA/ipoib: Set rtnl_link_ops for ipoib interfaces"), if the IPoIB device is moved to non-initial netns, destroying that netns lets the device vanish instead of moving it back to the initial netns, This is happening because default_device_exit() skips the interfaces due to having rtnl_link_ops set. Steps to reporoduce: ip netns add foo ip link set mlx5_ib0 netns foo ip netns delete foo WARNING: CPU: 1 PID: 704 at net/core/dev.c:11435 netdev_exit+0x3f/0x50 Modules linked in: xt_CHECKSUM xt_MASQUERADE xt_conntrack ipt_REJECT nf_reject_ipv4 nft_compat nft_counter nft_chain_nat nf_nat nf_conntrack nf_defrag_ipv6 nf_defrag_ipv4 nf_tables nfnetlink tun d fuse CPU: 1 PID: 704 Comm: kworker/u64:3 Tainted: G S W 5.13.0-rc1+ #1 Hardware name: Dell Inc. PowerEdge R630/02C2CP, BIOS 2.1.5 04/11/2016 Workqueue: netns cleanup_net RIP: 0010:netdev_exit+0x3f/0x50 Code: 48 8b bb 30 01 00 00 e8 ef 81 b1 ff 48 81 fb c0 3a 54 a1 74 13 48 8b 83 90 00 00 00 48 81 c3 90 00 00 00 48 39 d8 75 02 5b c3 <0f> 0b 5b c3 66 66 2e 0f 1f 84 00 00 00 00 00 66 90 0f 1f 44 00 RSP: 0018:ffffb297079d7e08 EFLAGS: 00010206 RAX: ffff8eb542c00040 RBX: ffff8eb541333150 RCX: 000000008010000d RDX: 000000008010000e RSI: 000000008010000d RDI: ffff8eb440042c00 RBP: ffffb297079d7e48 R08: 0000000000000001 R09: ffffffff9fdeac00 R10: ffff8eb5003be000 R11: 0000000000000001 R12: ffffffffa1545620 R13: ffffffffa1545628 R14: 0000000000000000 R15: ffffffffa1543b20 FS: 0000000000000000(0000) GS:ffff8ed37fa00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00005601b5f4c2e8 CR3: 0000001fc8c10002 CR4: 00000000003706e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: ops_exit_list.isra.9+0x36/0x70 cleanup_net+0x234/0x390 process_one_work+0x1cb/0x360 ? process_one_work+0x360/0x360 worker_thread+0x30/0x370 ? process_one_work+0x360/0x360 kthread+0x116/0x130 ? kthread_park+0x80/0x80 ret_from_fork+0x22/0x30 To avoid the above warning and later on the kernel panic that could happen on shutdown due to a NULL pointer dereference, make sure to set the netns_refund flag that was introduced by commit 3a5ca857079e ("can: dev: Move device back to init netns on owning netns delete") to properly restore the IPoIB interfaces to the initial netns. Fixes: 5ce2dced8e95 ("RDMA/ipoib: Set rtnl_link_ops for ipoib interfaces") Link: https://lore.kernel.org/r/20210525150134.139342-1-kamalheib1@gmail.com Signed-off-by: Kamal Heib <kamalheib1@gmail.com> Reviewed-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit a3e74fb9247cd530dca246699d5eb5a691884d32 upstream.

After the commit 5ce2dced8e95 ("RDMA/ipoib: Set rtnl_link_ops for ipoib
interfaces"), if the IPoIB device is moved to non-initial netns,
destroying that netns lets the device vanish instead of moving it back to
the initial netns, This is happening because default_device_exit() skips
the interfaces due to having rtnl_link_ops set.

Steps to reporoduce:
  ip netns add foo
  ip link set mlx5_ib0 netns foo
  ip netns delete foo

WARNING: CPU: 1 PID: 704 at net/core/dev.c:11435 netdev_exit+0x3f/0x50
Modules linked in: xt_CHECKSUM xt_MASQUERADE xt_conntrack ipt_REJECT
nf_reject_ipv4 nft_compat nft_counter nft_chain_nat nf_nat nf_conntrack
nf_defrag_ipv6 nf_defrag_ipv4 nf_tables nfnetlink tun d
 fuse
CPU: 1 PID: 704 Comm: kworker/u64:3 Tainted: G S      W  5.13.0-rc1+ #1
Hardware name: Dell Inc. PowerEdge R630/02C2CP, BIOS 2.1.5 04/11/2016
Workqueue: netns cleanup_net
RIP: 0010:netdev_exit+0x3f/0x50
Code: 48 8b bb 30 01 00 00 e8 ef 81 b1 ff 48 81 fb c0 3a 54 a1 74 13 48
8b 83 90 00 00 00 48 81 c3 90 00 00 00 48 39 d8 75 02 5b c3 <0f> 0b 5b
c3 66 66 2e 0f 1f 84 00 00 00 00 00 66 90 0f 1f 44 00
RSP: 0018:ffffb297079d7e08 EFLAGS: 00010206
RAX: ffff8eb542c00040 RBX: ffff8eb541333150 RCX: 000000008010000d
RDX: 000000008010000e RSI: 000000008010000d RDI: ffff8eb440042c00
RBP: ffffb297079d7e48 R08: 0000000000000001 R09: ffffffff9fdeac00
R10: ffff8eb5003be000 R11: 0000000000000001 R12: ffffffffa1545620
R13: ffffffffa1545628 R14: 0000000000000000 R15: ffffffffa1543b20
FS:  0000000000000000(0000) GS:ffff8ed37fa00000(0000) knlGS:0000000000000000
CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
CR2: 00005601b5f4c2e8 CR3: 0000001fc8c10002 CR4: 00000000003706e0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
Call Trace:
 ops_exit_list.isra.9+0x36/0x70
 cleanup_net+0x234/0x390
 process_one_work+0x1cb/0x360
 ? process_one_work+0x360/0x360
 worker_thread+0x30/0x370
 ? process_one_work+0x360/0x360
 kthread+0x116/0x130
 ? kthread_park+0x80/0x80
 ret_from_fork+0x22/0x30

To avoid the above warning and later on the kernel panic that could happen
on shutdown due to a NULL pointer dereference, make sure to set the
netns_refund flag that was introduced by commit 3a5ca857079e ("can: dev:
Move device back to init netns on owning netns delete") to properly
restore the IPoIB interfaces to the initial netns.

Fixes: 5ce2dced8e95 ("RDMA/ipoib: Set rtnl_link_ops for ipoib interfaces")
Link: https://lore.kernel.org/r/20210525150134.139342-1-kamalheib1@gmail.com
Signed-off-by: Kamal Heib <kamalheib1@gmail.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
RDMA/srpt: Fix error return code in srpt_cm_req_recv() 2021-05-14T07:44:28+00:00 Wang Wensheng wangwensheng4@huawei.com 2021-04-08T11:31:32+00:00 ed016b77012ee53a569c36b1a1d24e5e60f4672a [ Upstream commit 6bc950beff0c440ac567cdc4e7f4542a9920953d ] Fix to return a negative error code from the error handling case instead of 0, as done elsewhere in this function. Fixes: db7683d7deb2 ("IB/srpt: Fix login-related race conditions") Link: https://lore.kernel.org/r/20210408113132.87250-1-wangwensheng4@huawei.com Reported-by: Hulk Robot <hulkci@huawei.com> Signed-off-by: Wang Wensheng <wangwensheng4@huawei.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 6bc950beff0c440ac567cdc4e7f4542a9920953d ]

Fix to return a negative error code from the error handling case instead
of 0, as done elsewhere in this function.

Fixes: db7683d7deb2 ("IB/srpt: Fix login-related race conditions")
Link: https://lore.kernel.org/r/20210408113132.87250-1-wangwensheng4@huawei.com
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Wang Wensheng <wangwensheng4@huawei.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
RDMA/ipoib: Set rtnl_link_ops for ipoib interfaces 2020-10-29T08:57:51+00:00 Kamal Heib kamalheib1@gmail.com 2020-10-04T13:29:48+00:00 dc1d4c658b9c123e31054fffcbc0b23566694b1a [ Upstream commit 5ce2dced8e95e76ff7439863a118a053a7fc6f91 ] Report the "ipoib pkey", "mode" and "umcast" netlink attributes for every IPoiB interface type, not just children created with 'ip link add'. After setting the rtnl_link_ops for the parent interface, implement the dellink() callback to block users from trying to remove it. Fixes: 862096a8bbf8 ("IB/ipoib: Add more rtnl_link_ops callbacks") Link: https://lore.kernel.org/r/20201004132948.26669-1-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 5ce2dced8e95e76ff7439863a118a053a7fc6f91 ]

Report the "ipoib pkey", "mode" and "umcast" netlink attributes for every
IPoiB interface type, not just children created with 'ip link add'.

After setting the rtnl_link_ops for the parent interface, implement the
dellink() callback to block users from trying to remove it.

Fixes: 862096a8bbf8 ("IB/ipoib: Add more rtnl_link_ops callbacks")
Link: https://lore.kernel.org/r/20201004132948.26669-1-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>
IB/iser: Always check sig MR before putting it to the free pool 2020-10-01T11:17:50+00:00 Sergey Gorenko sergeygo@mellanox.com 2020-03-25T15:12:10+00:00 6e3b662d86a5e981c74bf88b8f4cdc3305040dfc [ Upstream commit 26e28deb813eed908cf31a6052870b6493ec0e86 ] libiscsi calls the check_protection transport handler only if SCSI-Respose is received. So, the handler is never called if iSCSI task is completed for some other reason like a timeout or error handling. And this behavior looks correct. But the iSER does not handle this case properly because it puts a non-checked signature MR to the free pool. Then the error occurs at reusing the MR because it is not allowed to invalidate a signature MR without checking. This commit adds an extra check to iser_unreg_mem_fastreg(), which is a part of the task cleanup flow. Now the signature MR is checked there if it is needed. Link: https://lore.kernel.org/r/20200325151210.1548-1-sergeygo@mellanox.com Signed-off-by: Sergey Gorenko <sergeygo@mellanox.com> Reviewed-by: Max Gurtovoy <maxg@mellanox.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 26e28deb813eed908cf31a6052870b6493ec0e86 ]

libiscsi calls the check_protection transport handler only if SCSI-Respose
is received. So, the handler is never called if iSCSI task is completed
for some other reason like a timeout or error handling. And this behavior
looks correct. But the iSER does not handle this case properly because it
puts a non-checked signature MR to the free pool. Then the error occurs at
reusing the MR because it is not allowed to invalidate a signature MR
without checking.

This commit adds an extra check to iser_unreg_mem_fastreg(), which is a
part of the task cleanup flow. Now the signature MR is checked there if it
is needed.

Link: https://lore.kernel.org/r/20200325151210.1548-1-sergeygo@mellanox.com
Signed-off-by: Sergey Gorenko <sergeygo@mellanox.com>
Reviewed-by: Max Gurtovoy <maxg@mellanox.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
IB/isert: Fix unaligned immediate-data handling 2020-09-17T11:47:44+00:00 Sagi Grimberg sagi@grimberg.me 2020-09-04T19:50:39+00:00 0f632bc483553fd2203face1384119d49edeff61 [ Upstream commit 0b089c1ef7047652b13b4cdfdb1e0e7dbdb8c9ab ] Currently we allocate rx buffers in a single contiguous buffers for headers (iser and iscsi) and data trailer. This means that most likely the data starting offset is aligned to 76 bytes (size of both headers). This worked fine for years, but at some point this broke, resulting in data corruptions in isert when a command comes with immediate data and the underlying backend device assumes 512 bytes buffer alignment. We assume a hard-requirement for all direct I/O buffers to be 512 bytes aligned. To fix this, we should avoid passing unaligned buffers for I/O. Instead, we allocate our recv buffers with some extra space such that we can have the data portion align to 512 byte boundary. This also means that we cannot reference headers or data using structure but rather accessors (as they may move based on alignment). Also, get rid of the wrong __packed annotation from iser_rx_desc as this has only harmful effects (not aligned to anything). This affects the rx descriptors for iscsi login and data plane. Fixes: 3d75ca0adef4 ("block: introduce multi-page bvec helpers") Link: https://lore.kernel.org/r/20200904195039.31687-1-sagi@grimberg.me Reported-by: Stephen Rust <srust@blockbridge.com> Tested-by: Doug Dumitru <doug@dumitru.com> Signed-off-by: Sagi Grimberg <sagi@grimberg.me> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 0b089c1ef7047652b13b4cdfdb1e0e7dbdb8c9ab ]

Currently we allocate rx buffers in a single contiguous buffers for
headers (iser and iscsi) and data trailer. This means that most likely the
data starting offset is aligned to 76 bytes (size of both headers).

This worked fine for years, but at some point this broke, resulting in
data corruptions in isert when a command comes with immediate data and the
underlying backend device assumes 512 bytes buffer alignment.

We assume a hard-requirement for all direct I/O buffers to be 512 bytes
aligned. To fix this, we should avoid passing unaligned buffers for I/O.

Instead, we allocate our recv buffers with some extra space such that we
can have the data portion align to 512 byte boundary. This also means that
we cannot reference headers or data using structure but rather
accessors (as they may move based on alignment). Also, get rid of the
wrong __packed annotation from iser_rx_desc as this has only harmful
effects (not aligned to anything).

This affects the rx descriptors for iscsi login and data plane.

Fixes: 3d75ca0adef4 ("block: introduce multi-page bvec helpers")
Link: https://lore.kernel.org/r/20200904195039.31687-1-sagi@grimberg.me
Reported-by: Stephen Rust <srust@blockbridge.com>
Tested-by: Doug Dumitru <doug@dumitru.com>
Signed-off-by: Sagi Grimberg <sagi@grimberg.me>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
RDMA/ipoib: Fix ABBA deadlock with ipoib_reap_ah() 2020-08-21T11:05:30+00:00 Jason Gunthorpe jgg@nvidia.com 2020-06-25T17:42:19+00:00 0f334b6684558658d1a92ba98695432223015c81 [ 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 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>
RDMA/ipoib: Return void from ipoib_ib_dev_stop() 2020-08-21T11:05:30+00:00 Kamal Heib kamalheib1@gmail.com 2020-06-23T10:52:36+00:00 5412efa6285af7c2cd20b1ecf3412468bc76afbe [ 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 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>
IB/ipoib: Fix double free of skb in case of multicast traffic in CM mode 2020-06-03T06:21:26+00:00 Valentine Fatiev valentinef@mellanox.com 2020-05-27T13:47:05+00:00 21358b3e77af10df4c7977f3d10d8da6ebbe5396 [ Upstream commit 1acba6a817852d4aa7916d5c4f2c82f702ee9224 ] When connected mode is set, and we have connected and datagram traffic in parallel, ipoib might crash with double free of datagram skb. The current mechanism assumes that the order in the completion queue is the same as the order of sent packets for all QPs. Order is kept only for specific QP, in case of mixed UD and CM traffic we have few QPs (one UD and few CM's) in parallel. The problem: ---------------------------------------------------------- Transmit queue: ----------------- UD skb pointer kept in queue itself, CM skb kept in spearate queue and uses transmit queue as a placeholder to count the number of total transmitted packets. 0 1 2 3 4 5 6 7 8 9 10 11 12 13 .........127 ------------------------------------------------------------ NL ud1 UD2 CM1 ud3 cm2 cm3 ud4 cm4 ud5 NL NL NL ........... ------------------------------------------------------------ ^ ^ tail head Completion queue (problematic scenario) - the order not the same as in the transmit queue: 1 2 3 4 5 6 7 8 9 ------------------------------------ ud1 CM1 UD2 ud3 cm2 cm3 ud4 cm4 ud5 ------------------------------------ 1. CM1 'wc' processing - skb freed in cm separate ring. - tx_tail of transmit queue increased although UD2 is not freed. Now driver assumes UD2 index is already freed and it could be used for new transmitted skb. 0 1 2 3 4 5 6 7 8 9 10 11 12 13 .........127 ------------------------------------------------------------ NL NL UD2 CM1 ud3 cm2 cm3 ud4 cm4 ud5 NL NL NL ........... ------------------------------------------------------------ ^ ^ ^ (Bad)tail head (Bad - Could be used for new SKB) In this case (due to heavy load) UD2 skb pointer could be replaced by new transmitted packet UD_NEW, as the driver assumes its free. At this point we will have to process two 'wc' with same index but we have only one pointer to free. During second attempt to free the same skb we will have NULL pointer exception. 2. UD2 'wc' processing - skb freed according the index we got from 'wc', but it was already overwritten by mistake. So actually the skb that was released is the skb of the new transmitted packet and not the original one. 3. UD_NEW 'wc' processing - attempt to free already freed skb. NUll pointer exception. The fix: ----------------------------------------------------------------------- The fix is to stop using the UD ring as a placeholder for CM packets, the cyclic ring variables tx_head and tx_tail will manage the UD tx_ring, a new cyclic variables global_tx_head and global_tx_tail are introduced for managing and counting the overall outstanding sent packets, then the send queue will be stopped and waken based on these variables only. Note that no locking is needed since global_tx_head is updated in the xmit flow and global_tx_tail is updated in the NAPI flow only. A previous attempt tried to use one variable to count the outstanding sent packets, but it did not work since xmit and NAPI flows can run at the same time and the counter will be updated wrongly. Thus, we use the same simple cyclic head and tail scheme that we have today for the UD tx_ring. Fixes: 2c104ea68350 ("IB/ipoib: Get rid of the tx_outstanding variable in all modes") Link: https://lore.kernel.org/r/20200527134705.480068-1-leon@kernel.org Signed-off-by: Valentine Fatiev <valentinef@mellanox.com> Signed-off-by: Alaa Hleihel <alaa@mellanox.com> Signed-off-by: Leon Romanovsky <leonro@mellanox.com> Acked-by: Doug Ledford <dledford@redhat.com> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> Signed-off-by: Sasha Levin <sashal@kernel.org> 
[ Upstream commit 1acba6a817852d4aa7916d5c4f2c82f702ee9224 ]

When connected mode is set, and we have connected and datagram traffic in
parallel, ipoib might crash with double free of datagram skb.

The current mechanism assumes that the order in the completion queue is
the same as the order of sent packets for all QPs. Order is kept only for
specific QP, in case of mixed UD and CM traffic we have few QPs (one UD and
few CM's) in parallel.

The problem:
----------------------------------------------------------

Transmit queue:
-----------------
UD skb pointer kept in queue itself, CM skb kept in spearate queue and
uses transmit queue as a placeholder to count the number of total
transmitted packets.

0   1   2   3   4  5  6  7  8   9  10  11 12 13 .........127
------------------------------------------------------------
NL ud1 UD2 CM1 ud3 cm2 cm3 ud4 cm4 ud5 NL NL NL ...........
------------------------------------------------------------
    ^                                  ^
   tail                               head

Completion queue (problematic scenario) - the order not the same as in
the transmit queue:

  1  2  3  4  5  6  7  8  9
------------------------------------
 ud1 CM1 UD2 ud3 cm2 cm3 ud4 cm4 ud5
------------------------------------

1. CM1 'wc' processing
   - skb freed in cm separate ring.
   - tx_tail of transmit queue increased although UD2 is not freed.
     Now driver assumes UD2 index is already freed and it could be used for
     new transmitted skb.

0   1   2   3   4  5  6  7  8   9  10  11 12 13 .........127
------------------------------------------------------------
NL NL  UD2 CM1 ud3 cm2 cm3 ud4 cm4 ud5 NL NL NL ...........
------------------------------------------------------------
        ^   ^                       ^
      (Bad)tail                    head
(Bad - Could be used for new SKB)

In this case (due to heavy load) UD2 skb pointer could be replaced by new
transmitted packet UD_NEW, as the driver assumes its free.  At this point
we will have to process two 'wc' with same index but we have only one
pointer to free.

During second attempt to free the same skb we will have NULL pointer
exception.

2. UD2 'wc' processing
   - skb freed according the index we got from 'wc', but it was already
     overwritten by mistake. So actually the skb that was released is the
     skb of the new transmitted packet and not the original one.

3. UD_NEW 'wc' processing
   - attempt to free already freed skb. NUll pointer exception.

The fix:
-----------------------------------------------------------------------

The fix is to stop using the UD ring as a placeholder for CM packets, the
cyclic ring variables tx_head and tx_tail will manage the UD tx_ring, a
new cyclic variables global_tx_head and global_tx_tail are introduced for
managing and counting the overall outstanding sent packets, then the send
queue will be stopped and waken based on these variables only.

Note that no locking is needed since global_tx_head is updated in the xmit
flow and global_tx_tail is updated in the NAPI flow only.  A previous
attempt tried to use one variable to count the outstanding sent packets,
but it did not work since xmit and NAPI flows can run at the same time and
the counter will be updated wrongly. Thus, we use the same simple cyclic
head and tail scheme that we have today for the UD tx_ring.

Fixes: 2c104ea68350 ("IB/ipoib: Get rid of the tx_outstanding variable in all modes")
Link: https://lore.kernel.org/r/20200527134705.480068-1-leon@kernel.org
Signed-off-by: Valentine Fatiev <valentinef@mellanox.com>
Signed-off-by: Alaa Hleihel <alaa@mellanox.com>
Signed-off-by: Leon Romanovsky <leonro@mellanox.com>
Acked-by: Doug Ledford <dledford@redhat.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Sasha Levin <sashal@kernel.org>
scsi: Revert "RDMA/isert: Fix a recently introduced regression related to logout" 2020-02-28T16:22:25+00:00 Bart Van Assche bvanassche@acm.org 2020-02-13T05:08:59+00:00 d92e714a463d4350b6b036ad4b198fd93894f1ca commit 76261ada16dcc3be610396a46d35acc3efbda682 upstream. Since commit 04060db41178 introduces soft lockups when toggling network interfaces, revert it. Link: https://marc.info/?l=target-devel&m=158157054906196 Cc: Rahul Kundu <rahul.kundu@chelsio.com> Cc: Mike Marciniszyn <mike.marciniszyn@intel.com> Cc: Sagi Grimberg <sagi@grimberg.me> Reported-by: Dakshaja Uppalapati <dakshaja@chelsio.com> Fixes: 04060db41178 ("scsi: RDMA/isert: Fix a recently introduced regression related to logout") Signed-off-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 76261ada16dcc3be610396a46d35acc3efbda682 upstream.

Since commit 04060db41178 introduces soft lockups when toggling network
interfaces, revert it.

Link: https://marc.info/?l=target-devel&m=158157054906196
Cc: Rahul Kundu <rahul.kundu@chelsio.com>
Cc: Mike Marciniszyn <mike.marciniszyn@intel.com>
Cc: Sagi Grimberg <sagi@grimberg.me>
Reported-by: Dakshaja Uppalapati <dakshaja@chelsio.com>
Fixes: 04060db41178 ("scsi: RDMA/isert: Fix a recently introduced regression related to logout")
Signed-off-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
IB/srp: Never use immediate data if it is disabled by a user 2020-02-14T21:34:07+00:00 Sergey Gorenko sergeygo@mellanox.com 2020-01-15T13:30:55+00:00 996dc3d50add6069079fe82d8e17a0050cb22131 commit 0fbb37dd82998b5c83355997b3bdba2806968ac7 upstream. Some SRP targets that do not support specification SRP-2, put the garbage to the reserved bits of the SRP login response. The problem was not detected for a long time because the SRP initiator ignored those bits. But now one of them is used as SRP_LOGIN_RSP_IMMED_SUPP. And it causes a critical error on the target when the initiator sends immediate data. The ib_srp module has a use_imm_date parameter to enable or disable immediate data manually. But it does not help in the above case, because use_imm_date is ignored at handling the SRP login response. The problem is definitely caused by a bug on the target side, but the initiator's behavior also does not look correct. The initiator should not use immediate data if use_imm_date is disabled by a user. This commit adds an additional checking of use_imm_date at the handling of SRP login response to avoid unexpected use of immediate data. Fixes: 882981f4a411 ("RDMA/srp: Add support for immediate data") Link: https://lore.kernel.org/r/20200115133055.30232-1-sergeygo@mellanox.com Signed-off-by: Sergey Gorenko <sergeygo@mellanox.com> Reviewed-by: Bart Van Assche <bvanassche@acm.org> Signed-off-by: Jason Gunthorpe <jgg@mellanox.com> Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org> 
commit 0fbb37dd82998b5c83355997b3bdba2806968ac7 upstream.

Some SRP targets that do not support specification SRP-2, put the garbage
to the reserved bits of the SRP login response.  The problem was not
detected for a long time because the SRP initiator ignored those bits. But
now one of them is used as SRP_LOGIN_RSP_IMMED_SUPP. And it causes a
critical error on the target when the initiator sends immediate data.

The ib_srp module has a use_imm_date parameter to enable or disable
immediate data manually. But it does not help in the above case, because
use_imm_date is ignored at handling the SRP login response. The problem is
definitely caused by a bug on the target side, but the initiator's
behavior also does not look correct.  The initiator should not use
immediate data if use_imm_date is disabled by a user.

This commit adds an additional checking of use_imm_date at the handling of
SRP login response to avoid unexpected use of immediate data.

Fixes: 882981f4a411 ("RDMA/srp: Add support for immediate data")
Link: https://lore.kernel.org/r/20200115133055.30232-1-sergeygo@mellanox.com
Signed-off-by: Sergey Gorenko <sergeygo@mellanox.com>
Reviewed-by: Bart Van Assche <bvanassche@acm.org>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>