RDMA/bnxt_re: Fix the sparse warnings

Fix the following warnings reported

drivers/infiniband/hw/bnxt_re/qplib_rcfw.c:909:27: warning: invalid assignment: |=
drivers/infiniband/hw/bnxt_re/qplib_rcfw.c:909:27: left side has type restricted __le16
drivers/infiniband/hw/bnxt_re/qplib_rcfw.c:909:27: right side has type unsigned long
...
drivers/infiniband/hw/bnxt_re/qplib_fp.c:1620:44: warning: invalid assignment: |=
drivers/infiniband/hw/bnxt_re/qplib_fp.c:1620:44: left side has type restricted __le64
drivers/infiniband/hw/bnxt_re/qplib_fp.c:1620:44: right side has type unsigned long long

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202312200537.HoNqPL5L-lkp@intel.com/
Fixes: 07f830ae4913 ("RDMA/bnxt_re: Adds MSN table capability for Gen P7 adapters")
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1703046717-8914-1-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Adds MSN table capability for Gen P7 adapters

GenP7 HW expects an MSN table instead of PSN table. Check
for the HW retransmission capability and populate the MSN
table if HW retansmission is supported.

Signed-off-by: Damodharam Ammepalli <damodharam.ammepalli@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1701946060-13931-7-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Support new 5760X P7 devices

Add basic support for 5760X P7 devices. Add new chip
revisions. The first version support is similar to
the existing P5 adapters. Extend the current support
for P5 adapters to P7 also.

Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1701946060-13931-2-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Refactor the queue index update

The queue index wrap around logic is based on power of 2 size depth.
All queues are created with power of 2 depth. This increases the
memory usage by the driver. This change is required for the next
patches that avoids the power of 2 depth requirement for each of
the queues.

Update the function that increments producer index and consumer
index during wrap around. Also, changes the index handling across
multiple functions.

Signed-off-by: Chandramohan Akula <chandramohan.akula@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1698069803-1787-2-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Fix the handling of control path response data

Flag that indicate control path command completion should be cleared
only after copying the command response data. As soon as the is_in_used
flag is clear, the waiting thread can proceed with wrong response
data. This wrong data is causing multiple issues like wrong lkey
used in data traffic and wrong AH Id etc.

Use a memory barrier to ensure that the response data
is copied and visible to the process waiting on a different
cpu core before clearing the is_in_used flag.

Clear the is_in_used after copying the command response.

Fixes: bcfee4ce3e01 ("RDMA/bnxt_re: remove redundant cmdq_bitmap")
Signed-off-by: Saravanan Vajravel <saravanan.vajravel@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1695199280-13520-2-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Fix kernel doc errors

Fix set of the following errors due to use of wrong kernel doc format
to describe function parameters:

drivers/infiniband/hw/bnxt_re/qplib_rcfw.c:68: warning: Function parameter or member 'rcfw'
not described in '__wait_for_resp'

Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202308180600.oOnkIAQV-lkp@intel.com/
Closes: https://lore.kernel.org/oe-kbuild-all/202308180401.iaj2ktTc-lkp@intel.com/
Closes: https://lore.kernel.org/oe-kbuild-all/202308180214.Lt9NAhbM-lkp@intel.com/
Closes: https://lore.kernel.org/oe-kbuild-all/202308180055.6zM4AK6V-lkp@intel.com/
Closes: https://lore.kernel.org/oe-kbuild-all/202308172136.ipx1wvs6-lkp@intel.com/
Link: https://lore.kernel.org/r/4b22c385f1b68590ace8f82f2985d14b20999432.1692539554.git.leon@kernel.org
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>

RDMA/bnxt_re: Remove unnecessary variable initializations

Remove unnecessary variable initializations.

Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1691052326-32143-7-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Fix the sideband buffer size handling for FW commands

bnxt_qplib_rcfw_alloc_sbuf allocates 24 bytes and it is better to fit
on stack variables. This way we can avoid unwanted kmalloc call.
Call dma_alloc_coherent directly instead of wrapper
bnxt_qplib_rcfw_alloc_sbuf.

Also, FW expects the side buffer needs to be aligned to
BNXT_QPLIB_CMDQE_UNITS(16B). So align the size to have the
extra padding bytes.

Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Hongguang Gao <hongguang.gao@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1691052326-32143-4-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Fix hang during driver unload

Driver unload hits a hang during stress testing of load/unload.

stack trace snippet -

tasklet_kill at ffffffff9aabb8b2
bnxt_qplib_nq_stop_irq at ffffffffc0a805fb [bnxt_re]
bnxt_qplib_disable_nq at ffffffffc0a80c5b [bnxt_re]
bnxt_re_dev_uninit at ffffffffc0a67d15 [bnxt_re]
bnxt_re_remove_device at ffffffffc0a6af1d [bnxt_re]

tasklet_kill can hang if the tasklet is scheduled after it is disabled.

Modified the sequences to disable the interrupt first and synchronize
irq before disabling the tasklet.

Fixes: 1ac5a4047975 ("RDMA/bnxt_re: Add bnxt_re RoCE driver")
Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1689322969-25402-3-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Refactor code around bnxt_qplib_map_rc()

Update function comment of bnxt_qplib_map_rc()

Remove intermediate return value ENXIO and directly called
bnxt_qplib_map_rc() from __send_message_basic_sanity().

Link: https://lore.kernel.org/r/20230616061700.741769-2-kashyap.desai@broadcom.com
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

RDMA/bnxt_re: Remove incorrect return check from slow path

The commit 691eb7c6110f ("RDMA/bnxt_re: handle command completions after
driver detect a timedout") introduced code resulting in below warning
issued by the smatch static checker.

drivers/infiniband/hw/bnxt_re/qplib_rcfw.c:513 __bnxt_qplib_rcfw_send_message()
warn: duplicate check 'rc' (previous on line 506)

Fix the warning by removing incorrect code block.

Fixes: 691eb7c6110f ("RDMA/bnxt_re: handle command completions after driver detect a timedout")
Link: https://lore.kernel.org/r/20230616061700.741769-1-kashyap.desai@broadcom.com
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

RDMA/bnxt_re: Initialize opcode while sending message

Fix compilation warning:

drivers/infiniband/hw/bnxt_re/qplib_rcfw.c:325:18:
error: variable 'opcode' is uninitialized when used here [-Werror,-Wuninitialized]
crsqe->opcode = opcode;
^~~~~~
drivers/infiniband/hw/bnxt_re/qplib_rcfw.c:291:11:
note: initialize the variable 'opcode' to silence this warning
u8 opcode;
^
= '\0'

Fixes: bcfee4ce3e01 ("RDMA/bnxt_re: remove redundant cmdq_bitmap")
Link: https://lore.kernel.org/r/6ad1e44be2b560986da6fdc6b68da606413e9026.1686644105.git.leonro@nvidia.com
Acked-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

RDMA/bnxt_re: optimize the parameters passed to helper functions

Avoid passing arguments like Opcode which can be retrieved from
bnxt_qplib_crsqe structure.

Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1686308514-11996-18-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: remove redundant cmdq_bitmap

cmdq_bitmap is used to derive the next available index in the CMDQ.
This is not required as the we can get the next index
using the existing bnxt_qplib_crsqe array.

Driver will use bnxt_qplib_crsqe array and flag is_in_used to
derive valid entries. is_in_used is replacement of cmdq_bitmap.
There is no change in the existing mechanism of the circular buffer
used to get index.

Added opcode field in bnxt_qplib_crsqe array so that it is easy to map
opcode associated with pending rcfw command.

Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1686308514-11996-17-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: use firmware provided max request timeout

Firmware provides max request timeout value as part of hwrm_ver_get
API. Driver gets the timeout from firmware and if that interface is
not available then fall back to hardcoded timeout value.
Also, Add a helper function to check the FW status.

Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1686308514-11996-16-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: cancel all control path command waiters upon error

When an error is detected in FW, wake up all the waiters as the
all of them need to be completed with timeout. Add the device
error state also as a wait condition.

Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1686308514-11996-15-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: post destroy_ah for delayed completion of AH creation

AH create may be called from interrpt context and driver has a special
timeout (8 sec) for this command. This is to avoid soft lockups when
the FW command takes more time. Driver returns -ETIMEOUT and fail
create AH, without waiting for actual completion from firmware.
When FW completion is received, use is_waiter_alive flag to avoid
a regular completion path.

If create_ah opcode is detected in completion path which does not have
waiter alive, driver will fetch ah_id from successful firmware
completion in the interrupt context and sends destroy_ah command
for same ah_id. This special post is done in quick manner using helper
function __send_message_no_waiter.

timeout_send is only used for debugging purposes.
If timeout_send value keeps incrementing, it indicates out of sync
active ah counter between driver and firmware. This is a limitation
but graceful handling is possible in future.

Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1686308514-11996-13-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Add firmware stall check detection

Every completion will update last_seen value in the unit of jiffies.
last_seen field will be used to know if firmware is alive and
is useful to detect firmware stall.

Non blocking interface __wait_for_resp will have logic to detect
firmware stall. After every 10 second interval if __wait_for_resp
has not received completion for a given command it will check for
firmware stall condition.

If current jiffies is greater than last_seen
jiffies + RCFW_FW_STALL_TIMEOUT_SEC * HZ, it is a firmware stall.

Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1686308514-11996-12-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: handle command completions after driver detect a timedout

If calling context detect command timeout, associated memory stored on
stack will not be valid. If firmware complete the same command later,
this causes incorrect memory access by driver.

Added is_waiter_alive to handle delayed completion by firmware.
is_waiter_alive is set and reset under command queue lock.

Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1686308514-11996-11-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: add helper function __poll_for_resp

This interface will be used if the driver has not enabled interrupt
and/or interrupt is disabled for a short period of time.
Completion is not possible from interrupt so this interface does
self-polling.

Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1686308514-11996-10-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Simplify the function that sends the FW commands

- Use __send_message_basic_sanity helper function.
- Do not retry posting same command if there is a queue full detection.
- ENXIO is used to indicate controller recovery.
- In the case of ERR_DEVICE_DETACHED state, the driver should not post
commands to the firmware, but also return fabricated written code.

Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1686308514-11996-9-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: use shadow qd while posting non blocking rcfw command

Whenever there is a fast path IO and create/destroy resources from
the slow path is happening in parallel, we may notice high latency
of slow path command completion.

Introduces a shadow queue depth to prevent the outstanding requests
to the FW. Driver will not allow more than #RCFW_CMD_NON_BLOCKING_SHADOW_QD
non-blocking commands to the Firmware.

Shadow queue depth is a soft limit only for non-blocking
commands. Blocking commands will be posted to the firmware
as long as there is a free slot.

Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1686308514-11996-8-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Avoid the command wait if firmware is inactive

Add a check to avoid waiting if driver already detects a
FW timeout. Return success for resource destroy in case
the device is detached. Add helper function to map timeout
error code to success.

Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1686308514-11996-7-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Enhance the existing functions that wait for FW responses

Use jiffies based timewait instead of counting iteration for
commands that block for FW response.

Also add a poll routine for control path commands. This is for
polling completion if the waiting commands timeout. This avoids cases
where the driver misses completion interrupts.

Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1686308514-11996-6-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: set fixed command queue depth

There is no need of setting max command queue entries based on
firmware version check.

Removing deperecated code.

Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1686308514-11996-5-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: remove virt_func check while creating RoCE FW channel

There is a common FW communication offset for both PF and VF.
Removed code around virt_fn check while creating FW channel.

Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1686308514-11996-4-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Avoid calling wake_up threads from spin_lock context

bnxt_qplib_service_creq can be called from interrupt or tasklet or
process context. So the function take irq variant of spin_lock.
But when wake_up is invoked with the lock held, it is putting the
calling context to sleep.

[exception RIP: __wake_up_common+190]
RIP: ffffffffb7539d7e RSP: ffffa73300207ad8 RFLAGS: 00000083
RAX: 0000000000000001 RBX: ffff91fa295f69b8 RCX: dead000000000200
RDX: ffffa733344af940 RSI: ffffa73336527940 RDI: ffffa73336527940
RBP: 000000000000001c R8: 0000000000000002 R9: 00000000000299c0
R10: 0000017230de82c5 R11: 0000000000000002 R12: ffffa73300207b28
R13: 0000000000000000 R14: ffffa733341bf928 R15: 0000000000000000
ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018

Call the wakeup after releasing the lock.

Fixes: 1ac5a4047975 ("RDMA/bnxt_re: Add bnxt_re RoCE driver")
Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1686308514-11996-3-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: wraparound mbox producer index

Driver is not handling the wraparound of the mbox producer index correctly.
Currently the wraparound happens once u32 max is reached.

Bit 31 of the producer index register is special and should be set
only once for the first command. Because the producer index overflow
setting bit31 after a long time, FW goes to initialization sequence
and this causes FW hang.

Fix is to wraparound the mbox producer index once it reaches u16 max.

Fixes: cee0c7bba486 ("RDMA/bnxt_re: Refactor command queue management code")
Fixes: 1ac5a4047975 ("RDMA/bnxt_re: Add bnxt_re RoCE driver")
Signed-off-by: Kashyap Desai <kashyap.desai@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1686308514-11996-2-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Use unique names while registering interrupts

bnxt_re currently uses the names "bnxt_qplib_creq" and "bnxt_qplib_nq-0"
while registering IRQs. There is no way to distinguish the IRQs of
different device ports when there are multiple IB devices registered.
This could make the scenarios worse where one want to pin IRQs of a device
port to certain CPUs.

Fixed the code to use unique names which has PCI BDF information while
registering interrupts like: "bnxt_re-nq-0@pci:0000:65:00.0" and
"bnxt_re-creq@pci:0000:65:00.1".

Fixes: 1ac5a4047975 ("RDMA/bnxt_re: Add bnxt_re RoCE driver")
Link: https://lore.kernel.org/r/1684478897-12247-4-git-send-email-selvin.xavier@broadcom.com
Reviewed-by: Bhargava Chenna Marreddy <bhargava.marreddy@broadcom.com>
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

RDMA/bnxt_re: Disable/kill tasklet only if it is enabled

When the ulp hook to start the IRQ fails because the rings are not
available, tasklets are not enabled. In this case when the driver is
unloaded, driver calls CREQ tasklet_kill. This causes an indefinite hang
as the tasklet is not enabled.

Driver shouldn't call tasklet_kill if it is not enabled. So using the
creq->requested and nq->requested flags to identify if both tasklets/irqs
are registered. Checking this flag while scheduling the tasklet from
ISR. Also, added a cleanup for disabling tasklet, in case request_irq
fails during start_irq.

Check for return value for bnxt_qplib_rcfw_start_irq and in case the
bnxt_qplib_rcfw_start_irq fails, return bnxt_re_start_irq without
attempting to start NQ IRQs.

Fixes: 1ac5a4047975 ("RDMA/bnxt_re: Add bnxt_re RoCE driver")
Link: https://lore.kernel.org/r/1684478897-12247-2-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Kalesh AP <kalesh-anakkur.purayil@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

RDMA/bnxt_re: Enable congestion control by default

Enable Congesion control by default. Issue FW command
enable the CC during driver load and disable it during
unload.

Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1680169540-10029-8-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDAM/bnxt_re: Use tlv apis while processing the slow path commands

Use the new TLV APIs for existing slow path commands. The TLV
APIs will be used to populate extended headers for some of the
Firmware commands, which will be introduced in the patches that
follow.

Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1680169540-10029-7-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Reduce number of argumets to control path command APIs

Reducing the number of arguments to bnxt_qplib_rcfw_send_message
by enclosing all its arguments into a command message structure.
Use the same struct while passing the command information to
send_message.

Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1680169540-10029-5-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Convert RCFW_CMD_PREP macro to static inline function

Convert RCFW_CMD_PREP macro to static inline function.
Also, remove the cmd_flags passed as none of the functions
are using it.

Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1680169540-10029-4-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Leon Romanovsky <leon@kernel.org>

RDMA/bnxt_re: Use bitmap_zalloc() when applicable

Use 'bitmap_zalloc()' to simplify code, improve the semantic and avoid
some open-coded arithmetic in allocator arguments.

Also change the corresponding 'kfree()' into 'bitmap_free()' to keep
consistency.

Link: https://lore.kernel.org/r/5c029daf43b92fdc27926fe8a98084843437c498.1637872888.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

RDMA/bnxt_re: Scan the whole bitmap when checking if "disabling RCFW with pending cmd-bit"

The 'cmdq->cmdq_bitmap' bitmap is 'rcfw->cmdq_depth' bits long. The size
stored in 'cmdq->bmap_size' is the size of the bitmap in bytes.

Remove this erroneous 'bmap_size' and use 'rcfw->cmdq_depth' directly in
'bnxt_qplib_disable_rcfw_channel()'. Otherwise some error messages may be
missing.

Other uses of 'cmdq_bitmap' already take into account 'rcfw->cmdq_depth'
directly.

Fixes: 1ac5a4047975 ("RDMA/bnxt_re: Add bnxt_re RoCE driver")
Link: https://lore.kernel.org/r/47ed717c3070a1d0f53e7b4c768a4fd11caf365d.1636707421.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Acked-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

RDMA/bnxt_re: Use GFP_KERNEL in non atomic context

Use GFP_KERNEL instead of GFP_ATOMIC while allocating control path
structures which will be only called from non atomic context

Link: https://lore.kernel.org/r/1631709163-2287-10-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

RDMA/bnxt_re: Reduce the delay in polling for hwrm command completion

Driver has 1ms delay between the polling for atomic command completion.
Polling immediately after issuing command usually doesn't report any
completions. So all commands in the blocking path needs two iterations. So
effectively 1ms spend on each command. HW requires much lesser time for
each command. So reduce the delay to 1us and increase the iteration count
to wait for the same time.

Link: https://lore.kernel.org/r/1631709163-2287-5-git-send-email-selvin.xavier@broadcom.com
Reviewed-by: Leon Romanovsky <leonro@nvidia.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

RDMA/bnxt_re: Move device to error state upon device crash

When the L2 driver detects a device crash or device undergone reset, it
invokes a stop callback to recover from error.

The current RoCE driver doesn't recover the device. So move the device to
error state and dispatch fatal events to all qps Release the MSIx vectors
to avoid a crash when L2 driver disables the MSIx. Also, check for the
device state to avoid posting further commands to the HW.

Link: https://lore.kernel.org/r/1615968942-30970-1-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Naresh Kumar PBS <nareshkumar.pbs@broadcom.com>
Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

RDMA/bnxt_re: Convert tasklets to use new tasklet_setup() API

In preparation for unconditionally passing the struct tasklet_struct
pointer to all tasklet callbacks, switch to using the new tasklet_setup()
and from_tasklet() to pass the tasklet pointer explicitly.

Link: https://lore.kernel.org/r/20200903060637.424458-2-allen.lkml@gmail.com
Signed-off-by: Romain Perier <romain.perier@gmail.com>
Signed-off-by: Allen Pais <allen.lkml@gmail.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

RDMA/bnxt_re: Fix the qp table indexing

qp->id can be a value outside the max number of qp. Indexing the qp table
with the id can cause out of bounds crash. So changing the qp table
indexing by (qp->id % max_qp -1).

Allocating one extra entry for QP1. Some adapters create one more than the
max_qp requested to accommodate QP1. If the qp->id is 1, store the
inforamtion in the last entry of the qp table.

Fixes: f218d67ef004 ("RDMA/bnxt_re: Allow posting when QPs are in error")
Link: https://lore.kernel.org/r/1598292876-26529-4-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@nvidia.com>

RDMA/bnxt_re: Simplify obtaining queue entry from hw ring

Restructring the data path and control path queue management code to
simplify the way a queue element is extracted from the hardware ring.

Introduced a new function which will give a pointer to the next ring item
depending upon the current cons/prod index in the hardware queue.

Further, there are hardcoding when size of queue entry is calculated,
replacing it with an inline function. This function would be easier to
expand if need going forward.

The code section to initialize the PSN search areas has also been
restructured and couple of functions has been added there.

Link: https://lore.kernel.org/r/1585851136-2316-4-git-send-email-devesh.sharma@broadcom.com
Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: Reduce device page size detection code

Getting rid of the repeated code in the driver when deciding on the page
size of the hardware ring memory. A new common function would translate
the ring page size into device specific page size.

Link: https://lore.kernel.org/r/1585851136-2316-2-git-send-email-devesh.sharma@broadcom.com
Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: Remove set but not used variables 'pg' and 'idx'

Fixes gcc '-Wunused-but-set-variable' warning:

drivers/infiniband/hw/bnxt_re/qplib_rcfw.c: In function '__send_message':
drivers/infiniband/hw/bnxt_re/qplib_rcfw.c:101:10: warning:
variable 'idx' set but not used [-Wunused-but-set-variable]
drivers/infiniband/hw/bnxt_re/qplib_rcfw.c:101:6: warning:
variable 'pg' set but not used [-Wunused-but-set-variable]

commit cee0c7bba486 ("RDMA/bnxt_re: Refactor command queue management
code") involved this, but not used.

Link: https://lore.kernel.org/r/20200227064900.92255-1-yuehaibing@huawei.com
Signed-off-by: YueHaibing <yuehaibing@huawei.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: Refactor doorbell management functions

Moving all the fast path doorbell functions at one place under
qplib_res.h. To pass doorbell record information a new structure
bnxt_qplib_db_info has been introduced. Every roce object holds an
instance of this structure and doorbell information is initialized during
resource creation.

When DB is rung only the current queue index is read from hardware ring
and rest of the data is taken from pre-initialized dbinfo structure.

Link: https://lore.kernel.org/r/1581786665-23705-8-git-send-email-devesh.sharma@broadcom.com
Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: Refactor command queue management code

Refactoring the command queue (rcfw) management code. A new data-structure
is introduced to describe the bar register. each object which deals with
mmio space should have a descriptor structure. This structure specifically
hold DB register information. Thus, slow path creq structure now hold a
bar register descriptor.

Further cleanup the rcfw structure to introduce the command queue context
and command response event queue context structures. Rest of the rcfw
related code has been touched to incorporate these three structures.

Link: https://lore.kernel.org/r/1581786665-23705-6-git-send-email-devesh.sharma@broadcom.com
Signed-off-by: Naresh Kumar PBS <nareshkumar.pbs@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: Refactor hardware queue memory allocation

At top level there are three major data structure addition. viz
bnxt_qplib_hwq_attr, bnxt_qplib_sg_info and bnxt_qplib_tqm_ctx

Intorduction of first data structure reduces the arguments list to
bnxt_re_alloc_init_hwq() function. There are changes all over the driver
code to incorporate this new structure. The caller needs to fill the
attribute data structure and pass to this function.

The second data structure is to pass memory region description
viz. sghead, page_size and page_shift. There are changes all over the
driver code to initialize bnxt_re_sg_info data structure. The new data
structure helps to reduce the argument list of __alloc_pbl() function
call.

Till now the TQM rings related members were not collected under any
specific data-structure making it hard to manage. The third data
sctructure bnxt_qplib_tqm_ctx is added to refactor the TQM queue
allocation and initialization.

Link: https://lore.kernel.org/r/1581786665-23705-4-git-send-email-devesh.sharma@broadcom.com
Signed-off-by: Naresh Kumar PBS <nareshkumar.pbs@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

remove ioremap_nocache and devm_ioremap_nocache

ioremap has provided non-cached semantics by default since the Linux 2.6
days, so remove the additional ioremap_nocache interface.

Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Arnd Bergmann <arnd@arndb.de>

RDMA/bnxt_re: Enable SRIOV VF support on Broadcom's 57500 adapter series

Broadcom's 575xx adapter series has support for SRIOV VFs. Making changes
to enable SRIOV VF support. There are two major area where changes are
done:

- Added new DB location for control-path and data-path DB ring

- New devices do not need to issue the sriov-config slow-path command
thus, skipping to call that firmware command.

For now enabling support for 64 RoCE VFs.

Link: https://lore.kernel.org/r/1570081715-14301-1-git-send-email-devesh.sharma@broadcom.com
Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: Fix stack-out-of-bounds in bnxt_qplib_rcfw_send_message

Driver copies FW commands to the HW queue as units of 16 bytes. Some
of the command structures are not exact multiple of 16. So while copying
the data from those structures, the stack out of bounds messages are
reported by KASAN. The following error is reported.

[ 1337.530155] ==================================================================
[ 1337.530277] BUG: KASAN: stack-out-of-bounds in bnxt_qplib_rcfw_send_message+0x40a/0x850 [bnxt_re]
[ 1337.530413] Read of size 16 at addr ffff888725477a48 by task rmmod/2785

[ 1337.530540] CPU: 5 PID: 2785 Comm: rmmod Tainted: G OE 5.2.0-rc6+ #75
[ 1337.530541] Hardware name: Dell Inc. PowerEdge R730/0599V5, BIOS 1.0.4 08/28/2014
[ 1337.530542] Call Trace:
[ 1337.530548] dump_stack+0x5b/0x90
[ 1337.530556] ? bnxt_qplib_rcfw_send_message+0x40a/0x850 [bnxt_re]
[ 1337.530560] print_address_description+0x65/0x22e
[ 1337.530568] ? bnxt_qplib_rcfw_send_message+0x40a/0x850 [bnxt_re]
[ 1337.530575] ? bnxt_qplib_rcfw_send_message+0x40a/0x850 [bnxt_re]
[ 1337.530577] __kasan_report.cold.3+0x37/0x77
[ 1337.530581] ? _raw_write_trylock+0x10/0xe0
[ 1337.530588] ? bnxt_qplib_rcfw_send_message+0x40a/0x850 [bnxt_re]
[ 1337.530590] kasan_report+0xe/0x20
[ 1337.530592] memcpy+0x1f/0x50
[ 1337.530600] bnxt_qplib_rcfw_send_message+0x40a/0x850 [bnxt_re]
[ 1337.530608] ? bnxt_qplib_creq_irq+0xa0/0xa0 [bnxt_re]
[ 1337.530611] ? xas_create+0x3aa/0x5f0
[ 1337.530613] ? xas_start+0x77/0x110
[ 1337.530615] ? xas_clear_mark+0x34/0xd0
[ 1337.530623] bnxt_qplib_free_mrw+0x104/0x1a0 [bnxt_re]
[ 1337.530631] ? bnxt_qplib_destroy_ah+0x110/0x110 [bnxt_re]
[ 1337.530633] ? bit_wait_io_timeout+0xc0/0xc0
[ 1337.530641] bnxt_re_dealloc_mw+0x2c/0x60 [bnxt_re]
[ 1337.530648] bnxt_re_destroy_fence_mr+0x77/0x1d0 [bnxt_re]
[ 1337.530655] bnxt_re_dealloc_pd+0x25/0x60 [bnxt_re]
[ 1337.530677] ib_dealloc_pd_user+0xbe/0xe0 [ib_core]
[ 1337.530683] srpt_remove_one+0x5de/0x690 [ib_srpt]
[ 1337.530689] ? __srpt_close_all_ch+0xc0/0xc0 [ib_srpt]
[ 1337.530692] ? xa_load+0x87/0xe0
...
[ 1337.530840] do_syscall_64+0x6d/0x1f0
[ 1337.530843] entry_SYSCALL_64_after_hwframe+0x44/0xa9
[ 1337.530845] RIP: 0033:0x7ff5b389035b
[ 1337.530848] Code: 73 01 c3 48 8b 0d 2d 0b 2c 00 f7 d8 64 89 01 48 83 c8 ff c3 66 2e 0f 1f 84 00 00 00 00 00 90 f3 0f 1e fa b8 b0 00 00 00 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 8b 0d fd 0a 2c 00 f7 d8 64 89 01 48
[ 1337.530849] RSP: 002b:00007fff83425c28 EFLAGS: 00000206 ORIG_RAX: 00000000000000b0
[ 1337.530852] RAX: ffffffffffffffda RBX: 00005596443e6750 RCX: 00007ff5b389035b
[ 1337.530853] RDX: 000000000000000a RSI: 0000000000000800 RDI: 00005596443e67b8
[ 1337.530854] RBP: 0000000000000000 R08: 00007fff83424ba1 R09: 0000000000000000
[ 1337.530856] R10: 00007ff5b3902960 R11: 0000000000000206 R12: 00007fff83425e50
[ 1337.530857] R13: 00007fff8342673c R14: 00005596443e6260 R15: 00005596443e6750

[ 1337.530885] The buggy address belongs to the page:
[ 1337.530962] page:ffffea001c951dc0 refcount:0 mapcount:0 mapping:0000000000000000 index:0x0
[ 1337.530964] flags: 0x57ffffc0000000()
[ 1337.530967] raw: 0057ffffc0000000 0000000000000000 ffffffff1c950101 0000000000000000
[ 1337.530970] raw: 0000000000000000 0000000000000000 00000000ffffffff 0000000000000000
[ 1337.530970] page dumped because: kasan: bad access detected

[ 1337.530996] Memory state around the buggy address:
[ 1337.531072] ffff888725477900: 00 00 00 00 f1 f1 f1 f1 00 00 00 00 00 f2 f2 f2
[ 1337.531180] ffff888725477980: 00 00 00 00 00 00 00 00 00 00 00 f1 f1 f1 f1 00
[ 1337.531288] >ffff888725477a00: 00 f2 f2 f2 f2 f2 f2 00 00 00 f2 00 00 00 00 00
[ 1337.531393] ^
[ 1337.531478] ffff888725477a80: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[ 1337.531585] ffff888725477b00: 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00
[ 1337.531691] ==================================================================

Fix this by passing the exact size of each FW command to
bnxt_qplib_rcfw_send_message as req->cmd_size. Before sending
the command to HW, modify the req->cmd_size to number of 16 byte units.

Fixes: 1ac5a4047975 ("RDMA/bnxt_re: Add bnxt_re RoCE driver")
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Link: https://lore.kernel.org/r/1566468170-489-1-git-send-email-selvin.xavier@broadcom.com
Signed-off-by: Doug Ledford <dledford@redhat.com>

RDMA/bnxt_re: Use correct sizing on buffers holding page DMA addresses

umem->nmap is used while allocating internal buffer for storing
page DMA addresses. This causes out of bounds array access while iterating
the umem DMA-mapped SGL with umem page combining as umem->nmap can be
less than number of system pages in umem.

Use ib_umem_num_pages() instead of umem->nmap to size the page array.
Add a new structure (bnxt_qplib_sg_info) to pass sglist, npages and nmap.

Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Shiraz Saleem <shiraz.saleem@intel.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: Skip backing store allocation for 57500 series

The backing store to keep HW context data structures is allocated and
initialized by L2 driver. For 57500 chip RoCE driver do not require to
allocate and initialize additional memory. Changing to skip duplicate
allocation and initialization for 57500 adapters. Driver continues as
before for older chips.

This patch also takes care of stats context memory alignment to 128
boundary, a requirement for 57500 series of chip. Older chips do not care
of alignment, thus the change is unconditional.

Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: Add 64bit doorbells for 57500 series

The new chip series has 64 bit doorbell for notification queues. Thus,
both control and data path event queues need new routines to write 64 bit
doorbell. Adding the same. There is new doorbell interface between the
chip and driver. Changing the chip specific data structure definitions.

Additional significant changes are listed below
- bnxt_re_net_ring_free/alloc takes a new argument
- bnxt_qplib_enable_nq and enable_rcfw uses new doorbell offset
for new chip.
- DB mapping for NQ and CREQ now maps 8 bytes.
- DBR_DBR_* macros renames to DBC_DBC_*
- store nq_db_offset in a 32bit data type.
- got rid of __iowrite64_copy, used writeq instead.
- changed the DB header initialization to simpler scheme.

Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: fix a size calculation

This is from static analysis not from testing. Depending on the value
of rcfw->cmdq_depth, then this might not cause an issue at runtime.

The BITS_TO_LONGS() macro tells us how many longs it take to hold a
bitmap. In other words, it divides by the number if bits per long and
rounds up. Then we want to take that number and multiple by
sizeof(long) to get the number of bytes to allocate.

The code here does the multiplication first so the rounding up is done
in the wrong place. So imagine we want to allocate 1 bit, then
"(1 * 8) / 64 = 1" when we round up. But it should be
"(1 / 64) * 8 = 8". In other words, because of the rounding difference
we might allocate up to "sizeof(long) - 1" bytes fewer than intended.

Fixes: 1ac5a4047975 ("RDMA/bnxt_re: Add bnxt_re RoCE driver")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Acked-By: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

cross-tree: phase out dma_zalloc_coherent()

We already need to zero out memory for dma_alloc_coherent(), as such
using dma_zalloc_coherent() is superflous. Phase it out.

This change was generated with the following Coccinelle SmPL patch:

@ replace_dma_zalloc_coherent @
expression dev, size, data, handle, flags;
@@

-dma_zalloc_coherent(dev, size, handle, flags)
+dma_alloc_coherent(dev, size, handle, flags)

Suggested-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Luis Chamberlain <mcgrof@kernel.org>
[hch: re-ran the script on the latest tree]
Signed-off-by: Christoph Hellwig <hch@lst.de>

RDMA/bnxt_re: Increase depth of control path command queue

Increasing the depth of control path command queue to 8K entries to handle
burst of commands. This feature needs support from FW and the driver/fw
compatibility is checked from the interface version number.

Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: Prevent driver crash due to NULL pointer in error message print

crsqe->resp would be NULL in case the host command timed out before
getting a response from HW. Check for NULL pointer to avoid a potential
crash while printing the error message.

Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: Drop L2 async events silently

In some FW versions, RoCE driver also receives an async notification which
was directed to L2 driver. RoCE driver does not handle this and print a
message to syslog. Drop these notifications silently.

Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: Avoid NULL check after accessing the pointer

This is reported by smatch check. rcfw->creq_bar_reg_iomem is accessed in
bnxt_qplib_rcfw_stop_irq and this variable check afterwards doesn't make
sense. Also, rcfw->creq_bar_reg_iomem will never be NULL. So Removing
this check.

Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
Fixes: 6e04b1035689 ("RDMA/bnxt_re: Fix broken RoCE driver due to recent L2 driver changes")
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: Fix recursive lock warning in debug kernel

Fix possible recursive lock warning. Its a false warning as the locks are
part of two differnt HW Queue data structure - cmdq and creq. Debug kernel
is throwing the following warning and stack trace.

[ 783.914967] ============================================
[ 783.914970] WARNING: possible recursive locking detected
[ 783.914973] 4.19.0-rc2+ #33 Not tainted
[ 783.914976] --------------------------------------------
[ 783.914979] swapper/2/0 is trying to acquire lock:
[ 783.914982] 000000002aa3949d (&(&hwq->lock)->rlock){..-.}, at: bnxt_qplib_service_creq+0x232/0x350 [bnxt_re]
[ 783.914999]
but task is already holding lock:
[ 783.915002] 00000000be73920d (&(&hwq->lock)->rlock){..-.}, at: bnxt_qplib_service_creq+0x2a/0x350 [bnxt_re]
[ 783.915013]
other info that might help us debug this:
[ 783.915016] Possible unsafe locking scenario:

[ 783.915019] CPU0
[ 783.915021] ----
[ 783.915034] lock(&(&hwq->lock)->rlock);
[ 783.915035] lock(&(&hwq->lock)->rlock);
[ 783.915037]
*** DEADLOCK ***

[ 783.915038] May be due to missing lock nesting notation

[ 783.915039] 1 lock held by swapper/2/0:
[ 783.915040] #0: 00000000be73920d (&(&hwq->lock)->rlock){..-.}, at: bnxt_qplib_service_creq+0x2a/0x350 [bnxt_re]
[ 783.915044]
stack backtrace:
[ 783.915046] CPU: 2 PID: 0 Comm: swapper/2 Not tainted 4.19.0-rc2+ #33
[ 783.915047] Hardware name: Dell Inc. PowerEdge R730/0599V5, BIOS 1.0.4 08/28/2014
[ 783.915048] Call Trace:
[ 783.915049] <IRQ>
[ 783.915054] dump_stack+0x90/0xe3
[ 783.915058] __lock_acquire+0x106c/0x1080
[ 783.915061] ? sched_clock+0x5/0x10
[ 783.915063] lock_acquire+0xbd/0x1a0
[ 783.915065] ? bnxt_qplib_service_creq+0x232/0x350 [bnxt_re]
[ 783.915069] _raw_spin_lock_irqsave+0x4a/0x90
[ 783.915071] ? bnxt_qplib_service_creq+0x232/0x350 [bnxt_re]
[ 783.915073] bnxt_qplib_service_creq+0x232/0x350 [bnxt_re]
[ 783.915078] tasklet_action_common.isra.17+0x197/0x1b0
[ 783.915081] __do_softirq+0xcb/0x3a6
[ 783.915084] irq_exit+0xe9/0x100
[ 783.915085] do_IRQ+0x6a/0x120
[ 783.915087] common_interrupt+0xf/0xf
[ 783.915088] </IRQ>

Use nested notation for the spin_lock to avoid this warning.

Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: QPLIB: Add and use #define dev_fmt(fmt) "QPLIB: " fmt

Consistently use the "QPLIB: " prefix for dev_<level> logging.

Miscellanea:

o Add missing newlines to avoid possible message interleaving
o Coalesce consecutive dev_<level> uses that emit a message header to
avoid < 80 column lengths and mistakenly output on multiple lines
o Reflow modified lines to use 80 columns where appropriate
o Consistently use "%s: " where __func__ is output
o QPLIB: is now always output immediately after the dev_<level> header

Signed-off-by: Joe Perches <joe@perches.com>
Acked-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: Fix broken RoCE driver due to recent L2 driver changes

The recent changes in Broadcom's ethernet driver(L2 driver) broke
RoCE functionality in terms of MSIx vector allocation and
de-allocation.

There is a possibility that L2 driver would initiate MSIx vector
reallocation depending upon the requests coming from administrator.
In such cases L2 driver needs to free up all the MSIx vectors
allocated previously and reallocate/initialize those.

If RoCE driver is loaded and reshuffling is attempted, there will be
kernel crashes because RoCE driver would still be holding the MSIx
vectors but L2 driver would attempt to free in-use vectors. Thus
leading to a kernel crash.

Making changes in roce driver to fix crashes described above.
As part of solution L2 driver tells RoCE driver to release
the MSIx vector whenever there is a need. When RoCE driver
get message it sync up with all the running tasklets and IRQ
handlers and releases the vectors. L2 driver send one more
message to RoCE driver to resume the MSIx vectors. L2 driver
guarantees that RoCE vector do not change during reshuffling.

Fixes: ec86f14ea506 ("bnxt_en: Add ULP calls to stop and restart IRQs.")
Fixes: 08654eb213a8 ("bnxt_en: Change IRQ assignment for RDMA driver.")
Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: Avoid Hard lockup during error CQE processing

Hitting the following hardlockup due to a race condition in
error CQE processing.

[26146.879798] bnxt_en 0000:04:00.0: QPLIB: FP: CQ Processed Req
[26146.886346] bnxt_en 0000:04:00.0: QPLIB: wr_id[1251] = 0x0 with status 0xa
[26156.350935] NMI watchdog: Watchdog detected hard LOCKUP on cpu 4
[26156.357470] Modules linked in: nfsd auth_rpcgss nfs_acl lockd grace
[26156.447957] CPU: 4 PID: 3413 Comm: kworker/4:1H Kdump: loaded
[26156.457994] Hardware name: Dell Inc. PowerEdge R430/0CN7X8,
[26156.466390] Workqueue: ib-comp-wq ib_cq_poll_work [ib_core]
[26156.472639] Call Trace:
[26156.475379] <NMI> [<ffffffff98d0d722>] dump_stack+0x19/0x1b
[26156.481833] [<ffffffff9873f775>] watchdog_overflow_callback+0x135/0x140
[26156.489341] [<ffffffff9877f237>] __perf_event_overflow+0x57/0x100
[26156.496256] [<ffffffff98787c24>] perf_event_overflow+0x14/0x20
[26156.502887] [<ffffffff9860a580>] intel_pmu_handle_irq+0x220/0x510
[26156.509813] [<ffffffff98d16031>] perf_event_nmi_handler+0x31/0x50
[26156.516738] [<ffffffff98d1790c>] nmi_handle.isra.0+0x8c/0x150
[26156.523273] [<ffffffff98d17be8>] do_nmi+0x218/0x460
[26156.528834] [<ffffffff98d16d79>] end_repeat_nmi+0x1e/0x7e
[26156.534980] [<ffffffff987089c0>] ? native_queued_spin_lock_slowpath+0x1d0/0x200
[26156.543268] [<ffffffff987089c0>] ? native_queued_spin_lock_slowpath+0x1d0/0x200
[26156.551556] [<ffffffff987089c0>] ? native_queued_spin_lock_slowpath+0x1d0/0x200
[26156.559842] <EOE> [<ffffffff98d083e4>] queued_spin_lock_slowpath+0xb/0xf
[26156.567555] [<ffffffff98d15690>] _raw_spin_lock+0x20/0x30
[26156.573696] [<ffffffffc08381a1>] bnxt_qplib_lock_buddy_cq+0x31/0x40 [bnxt_re]
[26156.581789] [<ffffffffc083bbaa>] bnxt_qplib_poll_cq+0x43a/0xf10 [bnxt_re]
[26156.589493] [<ffffffffc083239b>] bnxt_re_poll_cq+0x9b/0x760 [bnxt_re]

The issue happens if RQ poll_cq or SQ poll_cq or Async error event tries to
put the error QP in flush list. Since SQ and RQ of each error qp are added
to two different flush list, we need to protect it using locks of
corresponding CQs. Difference in order of acquiring the lock in
SQ poll_cq and RQ poll_cq can cause a hard lockup.

Revisits the locking strategy and removes the usage of qplib_cq.hwq.lock.
Instead of this lock, introduces qplib_cq.flush_lock to handle
addition/deletion of QPs in flush list. Also, always invoke the flush_lock
in order (SQ CQ lock first and then RQ CQ lock) to avoid any potential
deadlock.

Other than the poll_cq context, the movement of QP to/from flush list can
be done in modify_qp context or from an async error event from HW.
Synchronize these operations using the bnxt_re verbs layer CQ locks.
To achieve this, adds a call back to the HW abstraction layer(qplib) to
bnxt_re ib_verbs layer in case of async error event. Also, removes the
buddy cq functions as it is no longer required.

Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: Fix incorrect DB offset calculation

To support host systems with non 4K page size, l2_db_size shall be
calculated with 4096 instead of PAGE_SIZE. Also, supply the host page size
to FW during initialization.

Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Jason Gunthorpe <jgg@mellanox.com>

RDMA/bnxt_re: Add SRQ support for Broadcom adapters

Shared receive queue (SRQ) is defined as a pool of
receive buffers shared among multiple QPs which belong
to same protection domain in a given process context.
Use of SRQ reduces the memory foot print of IB applications.

Broadcom adapters support SRQ, adding code-changes to enable
shared receive queue.

Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>

RDMA/bnxt_re: Add support for query firmware version

The device now reports firmware version thus, removing
the hard coded values of the FW version string and
redundant fw_rev hook from sysfs. Adding code to query
firmware version from underlying device and report it
through the kernel verb to get firmware version string.

Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>

RDMA/bnxt_re: Add memory barriers when processing CQ/EQ entries

The code determines if the next ring entry is valid before proceeding
further to read the rest of the entry. The CPU can re-order and read
the rest of the entry first, possibly reading a stale entry, if DMA
of a new entry happens right after reading it.

Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>

bnxt_re: fix a crash in qp error event processing

In bnxt_qplib_process_qp_event(), for qp error events we look up the
qp-handle and pass it for further processing. But we don't check if the
handle is NULL. This could lead to a crash in the called functions when
that qp-handle is dereferenced, if the qp is destroyed in the meantime.
Fix this by checking for a valid qp-handle in that function.

Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>

bnxt_re: Fix incorrect usage of test_bit()

test_bit() takes a bit number while the 'flags' field in
struct bnxt_qplib_rcfw was using actual BIT position converted
values.
Fix this by assigning bit numbers and use consistent APIs
all the flag values.
Also logging a message in case of failure.

Thanks to Dan Carpenter for pointing this out.

Suggested-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>

RDMA/bnxt_re: Remove set-but-not-used variables

Signed-off-by: Bart Van Assche <bart.vanassche@wdc.com>
Cc: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>

bnxt_re: Stop issuing further cmds to FW once a cmd times out

Once a cmd to FW times out(after 20s) it is reasonable to
assume the FW or atleast the control path is dead.
No point issuing further cmds to the FW as each subsequent cmd
with another 20s timeout will cascade resulting in unnecessary
traces and/or NMI Lockups.

Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>

RDMA/bnxt_re: Allow posting when QPs are in error

This patch allows driver to post send and receive
requests on QPs which are in error state.

Instead of flushing the QP in the context of polling
error CQEs, the QPs will be added to a flush list
maintained per CQ. QP state is moved to error.
QP is added to flush list if the user moves it
to error state using modify_qp also. After polling the HW
CQ in poll_cq routine, this flush list is traversed
and driver completes work requests on each QP in the flush
list, till the budget expires. The QP is moved out of
flush list during QP destroy or during modify_QP to RESET.

When ULPs post Work Requests while QP is in error state,
driver will store the ULP data and then increment the
QP producer s/w index, without ringing doorbell. It then
schedules a worker to invoke the CQ handler since the
interrupts wont be generated from the HW for this request.

Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>

RDMA/bnxt_re: Fixing the Control path command and response handling

Fixing a concurrency issue with creq handling. Each caller
was given a globally managed crsq element, which was
accessed outside a lock. This could result in corruption,
if lot of applications are simultaneously issuing Control Path
commands. Now, each caller will provide its own response buffer
and the responses will be copied under a lock.
Also, Fixing the queue full condition check for the CMDQ.

As a part of these changes, the control path code is refactored
to remove the code replication in the response status checking.

Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>

RDMA/bnxt_re: Add bnxt_re RoCE driver

This patch introduces the RoCE driver for the Broadcom
NetXtreme-E 10/25/40/50G RoCE HCAs.

The RoCE driver is a two part driver that relies on the parent
bnxt_en NIC driver to operate. The changes needed in the bnxt_en
driver have already been incorporated via Dave Miller's net tree
into the mainline kernel.

The vendor official git repository for this driver is available
on github as:
https://github.com/Broadcom/linux-rdma-nxt/

Signed-off-by: Eddie Wai <eddie.wai@broadcom.com>
Signed-off-by: Devesh Sharma <devesh.sharma@broadcom.com>
Signed-off-by: Somnath Kotur <somnath.kotur@broadcom.com>
Signed-off-by: Sriharsha Basavapatna <sriharsha.basavapatna@broadcom.com>
Signed-off-by: Selvin Xavier <selvin.xavier@broadcom.com>
Reviewed-by: Leon Romanovsky <leonro@mellanox.com>
Signed-off-by: Doug Ledford <dledford@redhat.com>