Pull rdma fixes from Jason Gunthorpe:
"Many bug fixes in several drivers:
- Fix misuse of the DMA API in rtrs
- Several irdma issues: hung task due to SQ flushing, incorrect
capability reporting to userspace, improper error handling for MW
corners, touching an uninitialized SGL for during invalidation.
- hns was using the wrong page size limits for the HW, an incorrect
calculation of wqe_shift causing WQE corruption, and mis computed a
timer id.
- Fix a crash in SRP triggered by blktests
- Fix compiler errors by calling virt_to_page() with the proper type
in siw
- Userspace triggerable deadlock in ODP
- mlx5 could use the wrong profile due to some driver loading races,
counters were not working in some device configurations, and a
crash on error unwind"
* tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma:
RDMA/irdma: Report RNR NAK generation in device caps
RDMA/irdma: Use s/g array in post send only when its valid
RDMA/irdma: Return correct WC error for bind operation failure
RDMA/irdma: Return error on MR deregister CQP failure
RDMA/irdma: Report the correct max cqes from query device
MAINTAINERS: Update maintainers of HiSilicon RoCE
RDMA/mlx5: Fix UMR cleanup on error flow of driver init
RDMA/mlx5: Set local port to one when accessing counters
RDMA/mlx5: Rely on RoCE fw cap instead of devlink when setting profile
IB/core: Fix a nested dead lock as part of ODP flow
RDMA/siw: Pass a pointer to virt_to_page()
RDMA/srp: Set scmnd->result only when scmnd is not NULL
RDMA/hns: Remove the num_qpc_timer variable
RDMA/hns: Fix wrong fixed value of qp->rq.wqe_shift
RDMA/hns: Fix supported page size
RDMA/cma: Fix arguments order in net device validation
RDMA/irdma: Fix drain SQ hang with no completion
RDMA/rtrs-srv: Pass the correct number of entries for dma mapped SGL
RDMA/rtrs-clt: Use the right sg_cnt after ib_dma_map_sg
When the RDMA auxiliary driver probes, it sets its profile based on
devlink driverinit value. The latter might not be in sync with FW yet
(In case devlink reload is not performed), thus causing a mismatch
between RDMA driver and FW. This results in the following FW syndrome
when the RDMA driver tries to adjust RoCE state, which fails the probe:
"0xC1F678 | modify_nic_vport_context: roce_en set on a vport that
doesn't support roce"
To prevent this, select the PF profile based on FW RoCE capability
instead of relying on devlink driverinit value.
To provide backward compatibility of the RoCE disable feature, on older
FW's where roce_rw is not set (FW RoCE capability is read-only), keep
the current behavior e.g., rely on devlink driverinit value.
Fixes: fbfa97b4d79f ("net/mlx5: Disable roce at HCA level")
Reviewed-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Michael Guralnik <michaelgur@nvidia.com>
Reviewed-by: Saeed Mahameed <saeedm@nvidia.com>
Signed-off-by: Maher Sanalla <msanalla@nvidia.com>
Link: https://lore.kernel.org/r/cb34ce9a1df4a24c135cb804db87f7d2418bd6cc.1661763459.git.leonro@nvidia.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>
Add a lock_class_key per mlx5 device to avoid a false positive
"possible circular locking dependency" warning by lockdep, on flows
which lock more than one mlx5 device, such as adding SF.
kernel log:
======================================================
WARNING: possible circular locking dependency detected
5.19.0-rc8+ #2 Not tainted
------------------------------------------------------
kworker/u20:0/8 is trying to acquire lock:
ffff88812dfe0d98 (&dev->intf_state_mutex){+.+.}-{3:3}, at: mlx5_init_one+0x2e/0x490 [mlx5_core]
but task is already holding lock:
ffff888101aa7898 (&(¬ifier->n_head)->rwsem){++++}-{3:3}, at: blocking_notifier_call_chain+0x5a/0x130
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #1 (&(¬ifier->n_head)->rwsem){++++}-{3:3}:
down_write+0x90/0x150
blocking_notifier_chain_register+0x53/0xa0
mlx5_sf_table_init+0x369/0x4a0 [mlx5_core]
mlx5_init_one+0x261/0x490 [mlx5_core]
probe_one+0x430/0x680 [mlx5_core]
local_pci_probe+0xd6/0x170
work_for_cpu_fn+0x4e/0xa0
process_one_work+0x7c2/0x1340
worker_thread+0x6f6/0xec0
kthread+0x28f/0x330
ret_from_fork+0x1f/0x30
-> #0 (&dev->intf_state_mutex){+.+.}-{3:3}:
__lock_acquire+0x2fc7/0x6720
lock_acquire+0x1c1/0x550
__mutex_lock+0x12c/0x14b0
mlx5_init_one+0x2e/0x490 [mlx5_core]
mlx5_sf_dev_probe+0x29c/0x370 [mlx5_core]
auxiliary_bus_probe+0x9d/0xe0
really_probe+0x1e0/0xaa0
__driver_probe_device+0x219/0x480
driver_probe_device+0x49/0x130
__device_attach_driver+0x1b8/0x280
bus_for_each_drv+0x123/0x1a0
__device_attach+0x1a3/0x460
bus_probe_device+0x1a2/0x260
device_add+0x9b1/0x1b40
__auxiliary_device_add+0x88/0xc0
mlx5_sf_dev_state_change_handler+0x67e/0x9d0 [mlx5_core]
blocking_notifier_call_chain+0xd5/0x130
mlx5_vhca_state_work_handler+0x2b0/0x3f0 [mlx5_core]
process_one_work+0x7c2/0x1340
worker_thread+0x59d/0xec0
kthread+0x28f/0x330
ret_from_fork+0x1f/0x30
other info that might help us debug this:
Possible unsafe locking scenario:
CPU0 CPU1
---- ----
lock(&(¬ifier->n_head)->rwsem);
lock(&dev->intf_state_mutex);
lock(&(¬ifier->n_head)->rwsem);
lock(&dev->intf_state_mutex);
*** DEADLOCK ***
4 locks held by kworker/u20:0/8:
#0: ffff888150612938 ((wq_completion)mlx5_events){+.+.}-{0:0}, at: process_one_work+0x6e2/0x1340
#1: ffff888100cafdb8 ((work_completion)(&work->work)#3){+.+.}-{0:0}, at: process_one_work+0x70f/0x1340
#2: ffff888101aa7898 (&(¬ifier->n_head)->rwsem){++++}-{3:3}, at: blocking_notifier_call_chain+0x5a/0x130
#3: ffff88813682d0e8 (&dev->mutex){....}-{3:3}, at:__device_attach+0x76/0x460
stack backtrace:
CPU: 6 PID: 8 Comm: kworker/u20:0 Not tainted 5.19.0-rc8+
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS rel-1.13.0-0-gf21b5a4aeb02-prebuilt.qemu.org 04/01/2014
Workqueue: mlx5_events mlx5_vhca_state_work_handler [mlx5_core]
Call Trace:
<TASK>
dump_stack_lvl+0x57/0x7d
check_noncircular+0x278/0x300
? print_circular_bug+0x460/0x460
? lock_chain_count+0x20/0x20
? register_lock_class+0x1880/0x1880
__lock_acquire+0x2fc7/0x6720
? register_lock_class+0x1880/0x1880
? register_lock_class+0x1880/0x1880
lock_acquire+0x1c1/0x550
? mlx5_init_one+0x2e/0x490 [mlx5_core]
? lockdep_hardirqs_on_prepare+0x400/0x400
__mutex_lock+0x12c/0x14b0
? mlx5_init_one+0x2e/0x490 [mlx5_core]
? mlx5_init_one+0x2e/0x490 [mlx5_core]
? _raw_read_unlock+0x1f/0x30
? mutex_lock_io_nested+0x1320/0x1320
? __ioremap_caller.constprop.0+0x306/0x490
? mlx5_sf_dev_probe+0x269/0x370 [mlx5_core]
? iounmap+0x160/0x160
mlx5_init_one+0x2e/0x490 [mlx5_core]
mlx5_sf_dev_probe+0x29c/0x370 [mlx5_core]
? mlx5_sf_dev_remove+0x130/0x130 [mlx5_core]
auxiliary_bus_probe+0x9d/0xe0
really_probe+0x1e0/0xaa0
__driver_probe_device+0x219/0x480
? auxiliary_match_id+0xe9/0x140
driver_probe_device+0x49/0x130
__device_attach_driver+0x1b8/0x280
? driver_allows_async_probing+0x140/0x140
bus_for_each_drv+0x123/0x1a0
? bus_for_each_dev+0x1a0/0x1a0
? lockdep_hardirqs_on_prepare+0x286/0x400
? trace_hardirqs_on+0x2d/0x100
__device_attach+0x1a3/0x460
? device_driver_attach+0x1e0/0x1e0
? kobject_uevent_env+0x22d/0xf10
bus_probe_device+0x1a2/0x260
device_add+0x9b1/0x1b40
? dev_set_name+0xab/0xe0
? __fw_devlink_link_to_suppliers+0x260/0x260
? memset+0x20/0x40
? lockdep_init_map_type+0x21a/0x7d0
__auxiliary_device_add+0x88/0xc0
? auxiliary_device_init+0x86/0xa0
mlx5_sf_dev_state_change_handler+0x67e/0x9d0 [mlx5_core]
blocking_notifier_call_chain+0xd5/0x130
mlx5_vhca_state_work_handler+0x2b0/0x3f0 [mlx5_core]
? mlx5_vhca_event_arm+0x100/0x100 [mlx5_core]
? lock_downgrade+0x6e0/0x6e0
? lockdep_hardirqs_on_prepare+0x286/0x400
process_one_work+0x7c2/0x1340
? lockdep_hardirqs_on_prepare+0x400/0x400
? pwq_dec_nr_in_flight+0x230/0x230
? rwlock_bug.part.0+0x90/0x90
worker_thread+0x59d/0xec0
? process_one_work+0x1340/0x1340
kthread+0x28f/0x330
? kthread_complete_and_exit+0x20/0x20
ret_from_fork+0x1f/0x30
</TASK>
Fixes: 6a3273217469 ("net/mlx5: SF, Port function state change support")
Signed-off-by: Moshe Shemesh <moshe@nvidia.com>
Reviewed-by: Shay Drory <shayd@nvidia.com>
Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Pull rdma updates from Jason Gunthorpe: "This cycle we got a new RDMA driver "ERDMA" for the Alibaba cloud environment. Otherwise the changes are dominated by rxe fixes. There is another RDMA driver on the list that might get merged next cycle, 'MANA' for the Azure cloud environment. Summary: - Bug fixes and small features for irdma, hns, siw, qedr, hfi1, mlx5 - General spelling/grammer fixes - rdma cm can follow changes in neighbours for control packets - Significant amounts of rxe fixes and spec compliance changes - Use the modern NAPI API - Use the bitmap API instead of open coding - Performance improvements for rtrs - Add the ERDMA driver for Alibaba cloud - Fix a use after free bug in SRP" * tag 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rdma/rdma: (99 commits) RDMA/ib_srpt: Unify checking rdma_cm_id condition in srpt_cm_req_recv() RDMA/rxe: Fix error unwind in rxe_create_qp() RDMA/mlx5: Add missing check for return value in get namespace flow RDMA/rxe: Split qp state for requester and completer RDMA/rxe: Generate error completion for error requester QP state RDMA/rxe: Update wqe_index for each wqe error completion RDMA/srpt: Fix a use-after-free RDMA/srpt: Introduce a reference count in struct srpt_device RDMA/srpt: Duplicate port name members IB/qib: Fix repeated "in" within comments RDMA/erdma: Add driver to kernel build environment RDMA/erdma: Add the ABI definitions RDMA/erdma: Add the erdma module RDMA/erdma: Add connection management (CM) support RDMA/erdma: Add verbs implementation RDMA/erdma: Add verbs header file RDMA/erdma: Add event queue implementation RDMA/erdma: Add cmdq implementation RDMA/erdma: Add main include file RDMA/erdma: Add the hardware related definitions ...
After replacing the MR cache with an Mkey cache, rename the variables and functions to fit the new meaning. Link: https://lore.kernel.org/r/20220726071911.122765-6-michaelgur@nvidia.com Signed-off-by: Aharon Landau <aharonl@nvidia.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>
Use software VHCA id when it's supported by the firmware. A unique id is allocated upon mlx5_mdev_init() and freed upon mlx5_mdev_uninit(), as such it stays the same during the full life cycle of the device including upon health recovery if occurred. The conjunction of sw_vhca_id with sw_owner_id will be a global unique id per function which uses mlx5_core. The sw_vhca_id is set upon init_hca command and is used to specify the VHCA that the NIC vport is affiliated with. This functionality is needed upon migration of VM which is MPV based. (i.e. multi port device). Signed-off-by: Yishai Hadas <yishaih@nvidia.com> Reviewed-by: Mark Bloch <mbloch@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
As part of the flows invoked by mlx5_devlink_eswitch_mode_set() get to mlx5_rescan_drivers_locked() which can call mlx5e_probe()/mlx5e_remove and register/unregister mlx5e driver ports accordingly. This can lead to deadlock once mlx5_devlink_eswitch_mode_set() will use devlink lock. Use devl_port_register/unregister() instead of devlink_port_register/unregister() and add devlink instance locks in the driver paths to this function to have it locked while calling devl_ API function. If remove or probe were called by module init or module cleanup flows, need to lock devlink just before calling devl_port_register(), otherwise it is called by attach/detach or register/unregister flow and we can have the flow locked. Added flag to distinguish between these cases. This will be used by the downstream patch to invoke mlx5_devlink_eswitch_mode_set() with devlink locked. Signed-off-by: Moshe Shemesh <moshe@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
Added support for managing new type of ICM for devices that support sw_owner_v2. Signed-off-by: Erez Shitrit <erezsh@mellanox.com> Signed-off-by: Yevgeny Kliteynik <kliteyn@nvidia.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com> Acked-by: Saeed Mahameed <saeedm@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>
Pull vfio updates from Alex Williamson: - Improvements to mlx5 vfio-pci variant driver, including support for parallel migration per PF (Yishai Hadas) - Remove redundant iommu_present() check (Robin Murphy) - Ongoing refactoring to consolidate the VFIO driver facing API to use vfio_device (Jason Gunthorpe) - Use drvdata to store vfio_device among all vfio-pci and variant drivers (Jason Gunthorpe) - Remove redundant code now that IOMMU core manages group DMA ownership (Jason Gunthorpe) - Remove vfio_group from external API handling struct file ownership (Jason Gunthorpe) - Correct typo in uapi comments (Thomas Huth) - Fix coccicheck detected deadlock (Wan Jiabing) - Use rwsem to remove races and simplify code around container and kvm association to groups (Jason Gunthorpe) - Harden access to devices in low power states and use runtime PM to enable d3cold support for unused devices (Abhishek Sahu) - Fix dma_owner handling of fake IOMMU groups (Jason Gunthorpe) - Set driver_managed_dma on vfio-pci variant drivers (Jason Gunthorpe) - Pass KVM pointer directly rather than via notifier (Matthew Rosato) * tag 'vfio-v5.19-rc1' of https://github.com/awilliam/linux-vfio: (38 commits) vfio: remove VFIO_GROUP_NOTIFY_SET_KVM vfio/pci: Add driver_managed_dma to the new vfio_pci drivers vfio: Do not manipulate iommu dma_owner for fake iommu groups vfio/pci: Move the unused device into low power state with runtime PM vfio/pci: Virtualize PME related registers bits and initialize to zero vfio/pci: Change the PF power state to D0 before enabling VFs vfio/pci: Invalidate mmaps and block the access in D3hot power state vfio: Change struct vfio_group::container_users to a non-atomic int vfio: Simplify the life cycle of the group FD vfio: Fully lock struct vfio_group::container vfio: Split up vfio_group_get_device_fd() vfio: Change struct vfio_group::opened from an atomic to bool vfio: Add missing locking for struct vfio_group::kvm kvm/vfio: Fix potential deadlock problem in vfio include/uapi/linux/vfio.h: Fix trivial typo - _IORW should be _IOWR instead vfio/pci: Use the struct file as the handle not the vfio_group kvm/vfio: Remove vfio_group from kvm vfio: Change vfio_group_set_kvm() to vfio_file_set_kvm() vfio: Change vfio_external_check_extension() to vfio_file_enforced_coherent() vfio: Remove vfio_external_group_match_file() ...
Take the wrapper version which picks default node into a header file. This reduces the number of exported functions. Signed-off-by: Tariq Toukan <tariqt@nvidia.com> Reviewed-by: Moshe Shemesh <moshe@nvidia.com> Signed-off-by: Saeed Mahameed <saeedm@nvidia.com>