staging: gpib: Add ines GPIB driver

Driver for ines PCI/PCMCIA boards

Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20240918121908.19366-18-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gpib: Add hp82341x GPIB driver

Driver for old hp82341x boards

Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20240918121908.19366-17-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gpib: Add hp82335x GPIB driver

Driver for old hp82335x boards.

Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20240918121908.19366-16-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gpib: Add gpio bitbang GPIB driver

GPIO bitbang driver for Rasbberry Pi 2/3/4/5

Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20240918121908.19366-15-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gpib: Add Frank Mori Hess FPGA PCI GPIB driver

Driver for Frank Mori Hess' FPGA based PCI board

Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20240918121908.19366-14-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gpib: Add Fluke cda based cards GPIB driver

Driver for Fluke cda based cards

Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20240918121908.19366-13-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gpib: Add Computer Equipment Corporation GPIB driver

Driver for Computer Equipment Corporation and compatible boards.

Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20240918121908.19366-12-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gpib: Add Computer Boards GPIB driver

Driver for Computer Boards interface cards.

Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20240918121908.19366-11-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gpib: Add Agilent/Keysight 82357x USB GPIB driver

Driver for the Agilent/Keysight USB dongles.

Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20240918121908.19366-10-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gpib: Add HP/Agilent/Keysight 8235xx PCI GPIB driver

Driver for the HP/Agilent/Keysight 8235xx boards.

Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20240918121908.19366-9-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gpib: Add nec7210 GPIB chip driver

Low level Chip driver for NEC7210 and compatible based boards.

Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20240918121908.19366-8-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gpib: Add tms9914 GPIB chip driver

Low level Chip driver used on a number of boards.

Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20240918121908.19366-7-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gpib: Add GPIB common core driver

This is the common core driver that interfaces with the userland code
and creates the gpib device files.

Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20240918121908.19366-6-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gpib: Add user api include files

User api include files used by drivers and userland code.
The files are also distributed with the userland package.
Since these include files have been used by many applications we
had to keep the camelCase enums, typedefs and uint8_t declarations.

Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20240918121908.19366-5-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gpib: Add common include files for GPIB drivers

Common include files used only by the drivers.

Signed-off-by: Dave Penkler <dpenkler@gmail.com>
Link: https://lore.kernel.org/r/20240918121908.19366-4-dpenkler@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rts5208: Remove unused driver

Wei Wang from Realsil contributed this driver in 2011.

The following reasons lead to the removal:
- This driver generates maintenance workload
- Did not find minimal documentation on the web.
- No blog entries about anyone using the rts5208 and rts5288 during the
  last years.
- Did not find any device that may has it in and is still available on
  the market.

Link: https://lore.kernel.org/linux-staging/2024100943-shank-washed-a765@gregkh/T/#t
Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/20241009193250.6211-1-philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vchiq_core: Drop vchiq_pagelist.h

vchiq_pagelist.h only defines one struct and a couple of macros.
It can be merged with vchiq_core since all the pagelist related
function helpers are now in vchiq_core for bulk transfers.

Move the struct and related macros to vchiq_core header and drop
vchiq_pagelist.h.

Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20240919142130.1331495-4-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vchiq_core: Move bulk data functions in vchiq_core

Bulk transfers core logic lives in vchiq_core.c, hence move all
the preparatory bulk data allocation helpers to vchiq_core.c (from
vchiq_arm).

The discrepancy was noticed when vchiq_prepare_bulk_data() and
vchiq_complete_bulk() are being used vchiq_core.c but are defined
in vchiq_arm. Now that they are now confined to vchiq_core.c,
they can be made static and their signatures from vchiq_core header
can be dropped.

vchiq_prepare_bulk_data() and vchiq_complete_bulk() depends on
struct vchiq_pagelist_info, cleanup_pagelist(), free_pagelist() and
create_pagelist() hence they are pulled in from vchiq_arm as well,
as part of this commit.

No functional changes intended in this patch.

Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20240919142130.1331495-3-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vchiq_core: Move remote_event_signal() vchiq_core

The function remote_event_signal() is declared in vchiq_core.h while
defined in vchiq_arm.c and used only in vchiq_core.c. Move the
definition to vchiq_core.c as it is only used in this file.

Also convert it to static and drop the function signature from
vchiq_core.h header. BELL2 doorbell macro is also moved from vchiq_arm
to vchiq_core as part of this change.

No functional changes intended in this patch.

Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20240919142130.1331495-2-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vchiq_core: Drop retry loop on -EINTR

-EINTR is returned by vchiq_queue_message() on receiving a fatal
signal to the process. Since the process is deemed to be terminated
anyway, do not retry queuing with vchiq_queue_message() on -EINTR.

Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20240918163100.870596-7-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vchiq_arm: Do not retry bulk transfers on -EINTR

-EINTR is returned by various vchiq bulk transfer code paths
on receiving a fatal signal to the process. Since the process is
deemed to be terminated anyway, do not retry the bulk transfer
on -EINTR.

Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20240918163100.870596-6-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vchiq_core: Return -EINTR when bulk transfers are interrupted

Bulk transfers for various VCHIQ modes use mutex_lock_killable() and
wait_for_completion_killable() variations. Currently, -EAGAIN is
returned if these are interrupted by a fatal signal.

-EAGAIN may mislead the caller into thinking the operation can be
retried, while in reality, the process has received a fatal signal
and is terminating. Therefore, we should update the return value to
align with what these killable functions would return, specifically
-EINTR (Interrupted system call).

Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20240918163100.870596-5-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vchiq_core: Return -EINTR in queue_message() on interrupt

queue_message() uses mutex_lock_killable() and
wait_for_completion_killable() variations of locking and wait event
completions respectively. These functions return either 0 (on success)
or -EINTR, if interrupted by a fatal signal (as documented in the
kernel).

However, queue_message() is currently returning -EAGAIN if these
killable functions are interrupted by fatal signals. Bubbling up
-EAGAIN might give a sense to the caller, that the code path can
be re-tried however, in actual sense, a fatal signal has been
received by the process and the process is going away.

Hence, we should align the return value with what these killable
versions will return i.e. -EINTR (Interrupted system call).

Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20240918163100.870596-4-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vchiq_core: Return on all errors from queue_message()

In vchiq_connect_internal(), a MAKE_CONNECT message is queued
if the connection is disconnected, but only -EAGAIN error is
checked on the error path and returned.

However, queue_message() can fail with other errors as well hence,
vchiq_connect_internal() should return in those cases as well.

Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20240918163100.870596-3-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vchiq_core: Use killable wait completions for bulk transfers

commit f27e47bc6b8b ("staging: vchiq: use completions instead of
semaphores") introduced completions for events in vchiq interface.
It introduced _interruptible() version of completions for waiting
on events. However, it missed a subtle down_interruptible() macro
override in vchiq_killable.h, which used to mask most of the signals
and only interrupt on fatal ones.

The above issue was fixed in commit a772f116702e ("staging: vchiq: switch
to wait_for_completion_killable"). Given the override logic of
down_interruptible() that existed in vchiq_killable.h, that commit
fixed the completions with the correct variation i.e. killable() family
of functions.

However, commit a772f116702e ("staging: vchiq: switch to
wait_for_completion_killable") later got reverted [1] due to high CPU
load noticed by various downstream and upstream distributions [2].
Reverting the commit solved this problem but the root cause was never
diagonsed and the entire commit was reverted.

This patch brings back killable version of wait events but only for
bulk transfers and queue_message() transfer code paths.

The idea is to bring back killable versions for various event
completions in a phased manner so that we do not re-regress again as
noticed in [2]. Hence, no other wait events are converted from
interruptible -> killable in this patch.

Since the bulk transfers are no longer interruptible (but killable),
drop the "_interruptible" suffix from all vchiq_bulk_xfer_* functions.

[1]: commit 086efbabdc04 ("staging: vchiq: revert "switch to wait_for_completion_killable"")
[2]: https://patchwork.kernel.org/project/linux-arm-kernel/cover/20190509143137.31254-1-nsaenzjulienne@suse.de/

Signed-off-by: Umang Jain <umang.jain@ideasonboard.com>
Link: https://lore.kernel.org/r/20240918163100.870596-2-umang.jain@ideasonboard.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: gdm724x: fix returning -1 with return equivalent errors

As in the TODO file, use proper error codes from PM callbacks and init.

Signed-off-by: Rodrigo Gobbi <rodrigo.gobbi.7@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20241007211124.170540-1-rodrigo.gobbi.7@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8712: remove parentheses after &

Remove parentheses after & to fix checkpatch warning
Unnecessary parentheses.

Signed-off-by: Xingquan Liu <b1n@b1n.io>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/20241003070353.65998-1-b1n@b1n.io
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8712: use kmalloc_array

Adhere to Linux kernel coding style.

Reported by checkpatch:

WARNING: Prefer kmalloc_array over kmalloc with multiply
+       pxmitpriv->pxmitbuf = kmalloc(NR_XMITBUFF * sizeof(struct xmit_buf), GFP_ATOMIC);

Signed-off-by: Tudor Gheorghiu <tudor.reda@gmail.com>
Reviewed-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/ZvBy2lB_ok_OCmVI@redaops
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: Switch back to struct platform_driver::remove()

After commit 0edb555a65d1 ("platform: Make platform_driver::remove()
return void") .remove() is (again) the right callback to implement for
platform drivers.

Convert all staging drivers to use .remove(), with the eventual goal to
drop struct platform_driver::remove_new(). As .remove() and .remove_new()
have the same prototypes, conversion is done by just changing the structure
member name in the driver initializer.

Signed-off-by: Sergio Paracuellos <sergio.paracuellos@gmail.com>
Link: https://lore.kernel.org/r/20241001085751.282113-1-sergio.paracuellos@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: olpc_dcon: Drop explicit initialization of struct i2c_device_id::driver_data to 0

These drivers don't use the driver_data member of struct i2c_device_id,
so don't explicitly initialize this member.

This prepares putting driver_data in an anonymous union which requires
either no initialization or named designators. But it's also a nice
cleanup on its own.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@baylibre.com>
Link: https://lore.kernel.org/r/20240920153430.503212-16-u.kleine-koenig@baylibre.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: most: i2c: Drop explicit initialization of struct i2c_device_id::driver_data to 0

These drivers don't use the driver_data member of struct i2c_device_id,
so don't explicitly initialize this member.

This prepares putting driver_data in an anonymous union which requires
either no initialization or named designators. But it's also a nice
cleanup on its own.

While touching the initializer, also remove the comma after the sentinel
entry.

Signed-off-by: Uwe Kleine-König <u.kleine-koenig@baylibre.com>
Link: https://lore.kernel.org/r/20240920153430.503212-15-u.kleine-koenig@baylibre.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: s_uGetDataDuration: Fix declaration formatting

This patch fixes formatting of s_uGetDataDuration function declaration.

Signed-off-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Link: https://lore.kernel.org/r/20240918191959.51539-15-dominik.karol.piatkowski@protonmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: s_uGetDataDuration: Rename uNextPktTime variable

This patch renames uNextPktTime to next_pkt_time in s_uGetDataDuration
function in order to avoid using camel case.

Signed-off-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Link: https://lore.kernel.org/r/20240918191959.51539-14-dominik.karol.piatkowski@protonmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: s_uGetDataDuration: Rename uAckTime variable

This patch renames uAckTime to ack_time in s_uGetDataDuration function
in order to avoid using camel case.

Signed-off-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Link: https://lore.kernel.org/r/20240918191959.51539-13-dominik.karol.piatkowski@protonmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: s_uGetDataDuration: Rename bLastFrag variable

This patch renames bLastFrag to last_frag in s_uGetDataDuration function
in order to avoid using camel case.

Signed-off-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Link: https://lore.kernel.org/r/20240918191959.51539-12-dominik.karol.piatkowski@protonmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: s_uGetDataDuration: Rename byFBOption parameter

This patch renames byFBOption to fb_option in s_uGetDataDuration
function in order to avoid using camel case.

Signed-off-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Link: https://lore.kernel.org/r/20240918191959.51539-11-dominik.karol.piatkowski@protonmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: s_uGetDataDuration: Rename uMACfragNum parameter

This patch renames uMACfragNum to mac_frag_num in s_uGetDataDuration
function in order to avoid using camel case.

Signed-off-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Link: https://lore.kernel.org/r/20240918191959.51539-10-dominik.karol.piatkowski@protonmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: s_uGetDataDuration: Rename cbLastFragmentSize parameter

This patch renames cbLastFragmentSize to last_fragment_size in
s_uGetDataDuration function in order to avoid using camel case.

Signed-off-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Link: https://lore.kernel.org/r/20240918191959.51539-9-dominik.karol.piatkowski@protonmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: s_uGetDataDuration: Rename uFragIdx parameter

This patch renames uFragIdx to frag_idx in s_uGetDataDuration function in
order to avoid using camel case.

Signed-off-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Link: https://lore.kernel.org/r/20240918191959.51539-8-dominik.karol.piatkowski@protonmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: s_uGetDataDuration: Rename bNeedAck parameter

This patch renames bNeedAck to need_ack in s_uGetDataDuration function
in order to avoid using camel case.

Signed-off-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Link: https://lore.kernel.org/r/20240918191959.51539-7-dominik.karol.piatkowski@protonmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: s_uGetDataDuration: Rename wRate parameter

This patch renames wRate to rate in s_uGetDataDuration function in order
to avoid using camel case.

Signed-off-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Link: https://lore.kernel.org/r/20240918191959.51539-6-dominik.karol.piatkowski@protonmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: s_uGetDataDuration: Rename byPktType parameter

This patch renames byPktType to pkt_type in s_uGetDataDuration function
in order to avoid using camel case.

Signed-off-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Link: https://lore.kernel.org/r/20240918191959.51539-5-dominik.karol.piatkowski@protonmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: s_uGetDataDuration: Rename cbFrameLength parameter

This patch renames cbFrameLength to frame_length in s_uGetDataDuration
function in order to avoid using camel case.

Signed-off-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Link: https://lore.kernel.org/r/20240918191959.51539-4-dominik.karol.piatkowski@protonmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: s_uGetDataDuration: Rename byDurType parameter

This patch renames byDurType to dur_type in s_uGetDataDuration function
in order to avoid using camel case.

Signed-off-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Link: https://lore.kernel.org/r/20240918191959.51539-3-dominik.karol.piatkowski@protonmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: s_uGetDataDuration: Rename pDevice parameter

This patch renames pDevice to priv in s_uGetDataDuration function in
order to avoid using camel case.

Signed-off-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Link: https://lore.kernel.org/r/20240918191959.51539-2-dominik.karol.piatkowski@protonmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6655: rxtx.c: Fix too long lines in get_rtscts_time

This patch fixes the lines exceeding 100 columns in get_rtscts_time
function.

Signed-off-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Tested-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/20240918165052.30386-1-dominik.karol.piatkowski@protonmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: vt6656: Update maintainer in TODO

Commit ed394dbf5371b03a5335a7ba1973ba124c0ced3d replaced Forest Bond
with Philipp Hortmann as vt665X maintainer in MAINTAINERS, but
drivers/staging/vt6656/TODO was not changed, rendering it stale. This
patch fixes it.

Signed-off-by: Dominik Karol Piątkowski <dominik.karol.piatkowski@protonmail.com>
Link: https://lore.kernel.org/r/20240917171937.22801-1-dominik.karol.piatkowski@protonmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: core: rtw_cmd: Missing a blank line after declarations

Fix checkpatch: WARNING: Missing a blank line after declarations

Signed-off-by: Manuel Quintero F <sakunix@yahoo.com>
Link: https://lore.kernel.org/r/20241005233308.4520-1-sakunix@yahoo.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove function pointer disable_interrupt

Remove function pointer disable_interrupt and use
DisableInterrupt8723BSdio directly to increase readability.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/fee97dadc88bbdaebd82c99d0b6106d58315bd85.1727966761.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove function pointer enable_interrupt

Remove function pointer enable_interrupt and use
EnableInterrupt8723BSdio directly to increase readability.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/728827e155bdcd9951683e485d789d60bc203815.1727966761.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove function pointer read_adapter_info

Remove function pointer read_adapter_info and use
ReadAdapterInfo8723BS directly to increase readability.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/24eec4df528051fee3cf850308e009f114e14288.1727966761.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove function pointer intf_chip_configure

Remove function pointer intf_chip_configure and use
rtl8723bs_interface_configure directly to increase readability.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/d542f172438c333c015b87376a20645eeeae1b99.1727966761.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove function pointer init_default_value

Remove function pointer init_default_value and use
rtl8723bs_init_default_value directly to increase readability.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/d0bb58235d54d1c7e4806c5ea3a50dbf77c293e7.1727966761.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove function pointer read_chip_version

Remove function pointer read_chip_version and use
rtl8723b_read_chip_version directly to increase readability.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/79b26478a2493a1d7c27f8e88e0bec56a653d082.1727966761.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove function pointer dm_deinit

Remove function pointer dm_deinit as it is not linked to any function.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/bb03b1309e9aa6bae988c5b7003b4f925f5c7027.1727966761.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove function pointer dm_init

Remove function pointer dm_init and use rtl8723b_init_dm_priv directly to
increase readability.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/e25134aacd3f784300f527d7e367b9f0f066254a.1727966761.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove function pointer free_recv_priv

Remove function pointer free_recv_priv and use rtl8723bs_free_recv_priv
directly to increase readability.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/17f06e3495abea1f6ad2a2f8d4f4ff1f23bef654.1727966761.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove function pointer init_recv_priv

Remove function pointer init_recv_priv and use rtl8723bs_init_recv_priv
directly to increase readability.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/d9576e47920b045d702069fd3167e38d889412e7.1727966761.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove function pointer free_xmit_priv

Remove function pointer free_xmit_priv and use rtl8723bs_free_xmit_priv
directly to increase readability.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/01ee48c459ddda882c7616e6cf257d96429027c2.1727966761.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove function pointer init_xmit_priv

Remove function pointer init_xmit_priv and use rtl8723bs_init_xmit_priv
directly to increase readability.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/322d2412d89ae384365ec1d000bb0fc62128a261.1727966761.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove function pointer free_hal_data

Remove function pointer free_hal_data and function rtl8723b_free_hal_data
as it is dead code.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/455b7a4645e6652815020635a7b34e56c2b96423.1727966761.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove function pointer hal_deinit

Remove function pointer hal_deinit and use rtl8723bs_hal_deinit directly
to increase readability.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/b0acbbdc372e01baabd1d98f824bc2a3c6c4c600.1727966761.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove function pointer hal_init

Remove function pointer hal_init and use rtl8723bs_hal_init directly to
increase readability.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Link: https://lore.kernel.org/r/5c838981141aad1275cbcbe862ac7885de9bb8e9.1727966761.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Staging: rtl8723bs: hal: odm: removed unnecessary braces

Removed unnecessary braces.

Signed-off-by: Michael Harris <michaelharriscode@gmail.com>
Link: https://lore.kernel.org/r/20240928025336.55940-1-michaelharriscode@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove unused enum with first entry IFACE_PORT0

Remove unused enum with first entry IFACE_PORT0.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/ff820d921ba3e3cd777d76213f39d8a1ad93f7f9.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove constant result macro get_iface_type

Remove macro get_iface_type that returns always false to shorten code.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/d382321bdd67fdce0ec2357920f67b5dd81ef426.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove constant result macro is_primary_adapter

Remove macro is_primary_adapter that returns always true to shorten code.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/72673acf8b0ada07530b0cb3705cde4cda5e752b.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove unused function rtl8723bs_inirp_deinit

Remove unused function rtl8723bs_inirp_deinit and belonging unused
function pointer in struct hal_ops inirp_deinit.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/dc6d35602d44cc676bebbd6d84733ea5420ac3f3.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove unused function rtl8723bs_inirp_init

Remove unused function rtl8723bs_inirp_init and belonging unused
function pointer in struct hal_ops inirp_init.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/617f5fe59abf5f44c70566db60cc624e304c678f.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove unused function GetHalODMVar

Remove unused function GetHalODMVar.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/f6f589189a72a55d3a57bd37299929c307f31b4f.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove unused function rtl8723b_GetHalODMVar

Remove unused function rtl8723b_GetHalODMVar and belonging unused
function pointer in struct hal_ops GetHalODMVarHandler.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/3837617badf7c81b2914074b56c5064276eb1946.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove unused function Hal_BT_EfusePowerSwitch

Remove unused function Hal_BT_EfusePowerSwitch and belonging unused
function pointer in struct hal_ops BTEfusePowerSwitch.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/2768736f969eaf935df1492ffd5afd98b05db11e.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove unused function PHY_GetTxPowerLevel8723B

Remove unused function PHY_GetTxPowerLevel8723B and belonging unused
function pointer in struct hal_ops get_tx_power_level_handler.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/65e86a58b513c580325fe93cc47a114f51437eea.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove unused function PHY_SetBWMode8723B

Remove unused function PHY_SetBWMode8723B and belonging unused function
pointer in struct hal_ops set_bwmode_handler.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/4f3a76bce3dc309a179e588d184765e54816d3d9.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove unused entries from struct hal_ops

Remove unused function pointers from struct hal_ops.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/d4c1f7b6765ec246c797f4d0ac4d429fe6826180.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove unused function _ReadCAM

Remove unused function _ReadCAM.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/f8ce4d176c2aa1d312183263658c4683a23a1e4c.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove unused function rtw_get_oper_bw

Remove unused function rtw_get_oper_bw.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/5918320008abc9a14c00fd61f00b40f35f1a5bef.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove unused function rtw_get_oper_choffset

Remove unused function rtw_get_oper_choffset.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/23623deed7bb225c614693d9b871e2d6f49744a0.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove unused function read_cam

Remove unused function read_cam.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/a4d6cce804f23d4ac8267a572d168356bc7e84ed.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove unused function rtw_search_max_mac_id

Remove unused function rtw_search_max_mac_id.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/34c5f2ef44641c5151dde12b161d3f0aa963de5c.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: rtl8723bs: Remove unused function dvobj_get_port0_adapter

Remove unused function dvobj_get_port0_adapter.

Signed-off-by: Philipp Hortmann <philipp.g.hortmann@gmail.com>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/r/11091c00a57600a79a623f92ca8435034f0dfb3c.1726339782.git.philipp.g.hortmann@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

staging: octeon: Use new initialization api for tasklet

Use the new api DECLARE_TASKLET instead of DECLARE_TASKLET_OLD
introduced in commit 12cc923f1ccc ("tasklet: Introduce new
initialization API").

This change updates the tasklet initialization
process without introducing any functional changes,
ensuring the code aligns with the new API.

Signed-off-by: Abhishek Tamboli <abhishektamboli9@gmail.com>
Link: https://lore.kernel.org/r/20240914184935.848999-1-abhishektamboli9@gmail.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Merge wireless-next into staging-next

This merges up to commit a0efa2f362a6 ("Merge net-next/main to resolve
conflicts") from wireless-next into the staging-next branch to handle
the removal of a staging driver easier (it happened in the wireless-next
branch).  That way we don't have developers confused and start
submitting changes for this driver that is now deleted.

Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

Merge net-next/main to resolve conflicts

The wireless-next tree was based on something older, and there
are now conflicts between -rc2 and work here. Merge net-next,
which has enough of -rc2 for the conflicts to happen, resolving
them in the process.

Signed-off-by: Johannes Berg <johannes.berg@intel.com>

Revert "wifi: cfg80211: unexport wireless_nlevent_flush()"

Revert this, I neglected to take into account the fact that
cfg80211 itself can be a module, but wext is always builtin.

Fixes: aee809aaa2d1 ("wifi: cfg80211: unexport wireless_nlevent_flush()")
Signed-off-by: Johannes Berg <johannes.berg@intel.com>

net: phy: microchip_t1: SQI support for LAN887x

Add support for measuring Signal Quality Index for LAN887x T1 PHY.
Signal Quality Index (SQI) is measure of Link Channel Quality from
0 to 7, with 7 as the best. By default, a link loss event shall
indicate an SQI of 0.

Signed-off-by: Tarun Alle <Tarun.Alle@microchip.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20241007063943.3233-1-tarun.alle@microchip.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'net-phy-marvell-88q2xxx-enable-auto-negotiation-for-mv88q2110'

Niklas Söderlund says:

====================
net: phy: marvell-88q2xxx: Enable auto negotiation for mv88q2110

This series enables auto negotiation for the mv88q2110 device.
Previously this feature have been disabled for mv88q2110, while enabled
for other devices supported by this driver.

The initial driver implementation states this is due to the
configuration sequence provided by the vendor did not work. By comparing
the initialization sequence of other devices this driver supports and
the out-of-tree PHY driver for mv88q2110 found in the Renesas BSP [1]
I was able to figure out a working configuration.

As I have no access to the datasheets of either of these devices it
would be super if someone who has could sanity check the initialization
sequence.

With this series I'm able to auto negotiate both 1000Mbps and 100Mbps
links without issue.

    # ethtool eth0
    Settings for eth0:
            Supported ports: [  ]
            Supported link modes:   100baseT1/Full
                                    1000baseT1/Full
            Supported pause frame use: Symmetric Receive-only
            Supports auto-negotiation: Yes
            Supported FEC modes: Not reported
            Advertised link modes:  100baseT1/Full
                                    1000baseT1/Full
            Advertised pause frame use: No
            Advertised auto-negotiation: Yes
            Advertised FEC modes: Not reported
            Link partner advertised link modes:  100baseT1/Full
                                                 1000baseT1/Full
            Link partner advertised pause frame use: No
            Link partner advertised auto-negotiation: Yes
            Link partner advertised FEC modes: Not reported
            Speed: 1000Mb/s
            Duplex: Full
            Auto-negotiation: on
            master-slave cfg: preferred master
            master-slave status: slave
            Port: Twisted Pair
            PHYAD: 0
            Transceiver: external
            MDI-X: Unknown
            Link detected: yes
            SQI: 15/15

And the performance is good too. Without this change I was not able to
manually configure a 1000Mbps link, only 100Mbps ones. So this gives a
huge performance boost for my use-case.

    [  5] local 10.1.0.2 port 5201 connected to 10.1.0.1 port 38346
    [ ID] Interval           Transfer     Bitrate         Retr  Cwnd
    [  5]   0.00-1.00   sec  96.8 MBytes   812 Mbits/sec    0    469 KBytes
    [  5]   1.00-2.00   sec  94.3 MBytes   791 Mbits/sec    0    469 KBytes
    [  5]   2.00-3.00   sec  96.1 MBytes   806 Mbits/sec    0    469 KBytes
    [  5]   3.00-4.00   sec  98.3 MBytes   825 Mbits/sec    0    469 KBytes
    [  5]   4.00-5.00   sec  98.4 MBytes   825 Mbits/sec    0    469 KBytes
    [  5]   5.00-6.00   sec  98.4 MBytes   826 Mbits/sec    0    469 KBytes
    [  5]   6.00-7.00   sec  98.9 MBytes   830 Mbits/sec    0    469 KBytes
    [  5]   7.00-8.00   sec  91.7 MBytes   769 Mbits/sec    0    469 KBytes
    [  5]   8.00-9.00   sec  99.4 MBytes   834 Mbits/sec    0    747 KBytes
    [  5]   9.00-10.00  sec   101 MBytes   851 Mbits/sec    0    747 KBytes

Patch 1/3 and 2/3 are preparation patches that align and move functions
around as the mv88q2110 code paths can now reuses much of what is done
for mv88q2220. While patch 3/3 adds the new initialization sequence and
removes the auto negotiation limit for mv88q2110.

1.  https://github.com/renesas-rcar/linux-bsp/commit/2a1f07d0e722a18188cfe62842b61f2fbc0ba812
====================

Link: https://patch.msgid.link/20241005112412.544360-1-niklas.soderlund+renesas@ragnatech.se
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: phy: marvell-88q2xxx: Enable auto negotiation for mv88q2110

The initial marvell-88q2xxx driver only supported the Marvell 88Q2110
PHY without auto negotiation support. The reason documented states that
the provided initialization sequence did not to work. Now a method to
enable auto negotiation have been found by comparing the initialization
of other supported devices and an out-of-tree PHY driver.

Perform the minimal needed initialization of the PHY to get auto
negotiation working and remove the limitation that disables the auto
negotiation feature for the mv88q2110 device.

With this change a 1000Mbps full duplex link is able to be negotiated
between two mv88q2110 and the link works perfectly. The other side also
reflects the manually configure settings of the master device.

    # ethtool eth0
    Settings for eth0:
            Supported ports: [  ]
            Supported link modes:   100baseT1/Full
                                    1000baseT1/Full
            Supported pause frame use: Symmetric Receive-only
            Supports auto-negotiation: Yes
            Supported FEC modes: Not reported
            Advertised link modes:  100baseT1/Full
                                    1000baseT1/Full
            Advertised pause frame use: No
            Advertised auto-negotiation: Yes
            Advertised FEC modes: Not reported
            Link partner advertised link modes:  100baseT1/Full
                                                 1000baseT1/Full
            Link partner advertised pause frame use: No
            Link partner advertised auto-negotiation: Yes
            Link partner advertised FEC modes: Not reported
            Speed: 1000Mb/s
            Duplex: Full
            Auto-negotiation: on
            master-slave cfg: preferred master
            master-slave status: slave
            Port: Twisted Pair
            PHYAD: 0
            Transceiver: external
            MDI-X: Unknown
            Link detected: yes
            SQI: 15/15

Before this change I was not able to manually configure 1000Mbps link,
only a 100Mpps link so this change providers an improvement in
performance for this device.

    [  5] local 10.1.0.2 port 5201 connected to 10.1.0.1 port 38346
    [ ID] Interval           Transfer     Bitrate         Retr  Cwnd
    [  5]   0.00-1.00   sec  96.8 MBytes   812 Mbits/sec    0    469 KBytes
    [  5]   1.00-2.00   sec  94.3 MBytes   791 Mbits/sec    0    469 KBytes
    [  5]   2.00-3.00   sec  96.1 MBytes   806 Mbits/sec    0    469 KBytes
    [  5]   3.00-4.00   sec  98.3 MBytes   825 Mbits/sec    0    469 KBytes
    [  5]   4.00-5.00   sec  98.4 MBytes   825 Mbits/sec    0    469 KBytes
    [  5]   5.00-6.00   sec  98.4 MBytes   826 Mbits/sec    0    469 KBytes
    [  5]   6.00-7.00   sec  98.9 MBytes   830 Mbits/sec    0    469 KBytes
    [  5]   7.00-8.00   sec  91.7 MBytes   769 Mbits/sec    0    469 KBytes
    [  5]   8.00-9.00   sec  99.4 MBytes   834 Mbits/sec    0    747 KBytes
    [  5]   9.00-10.00  sec   101 MBytes   851 Mbits/sec    0    747 KBytes

Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Tested-by: Stefan Eichenberger <eichest@gmail.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20241005112412.544360-4-niklas.soderlund+renesas@ragnatech.se
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: phy: marvell-88q2xxx: Make register writer function generic

In preparation to adding auto negotiation support to mv88q2110 move and
rename the helper function used to write an array of register values to
the PHY.

Just as for mv88q2220 devices this helper will be needed to for the
initial configuration of the mv88q2110 to support auto negotiation.

The function is moved verbatim, there is no change in behavior.

Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Tested-by: Dimitri Fedrau <dima.fedrau@gmail.com>
Tested-by: Stefan Eichenberger <eichest@gmail.com>
Link: https://patch.msgid.link/20241005112412.544360-3-niklas.soderlund+renesas@ragnatech.se
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: phy: marvell-88q2xxx: Align soft reset for mv88q2110 and mv88q2220

The soft reset implementations for mv88q2110 and mv88q2220 differ as the
later need to consider that auto negation is supported on mv88q2220
devices. In preparation of enabling auto negotiation on mv88q2110 merge
the two rest functions into a device generic one.

The mv88q2220 behavior is kept as is but extended to wait for the reset
bit to be clears before continuing, as was done previously on mv88q2220.

Signed-off-by: Niklas Söderlund <niklas.soderlund+renesas@ragnatech.se>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Tested-by: Dimitri Fedrau <dima.fedrau@gmail.com>
Tested-by: Stefan Eichenberger <eichest@gmail.com>
Link: https://patch.msgid.link/20241005112412.544360-2-niklas.soderlund+renesas@ragnatech.se
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

fsl/fman: Fix a typo

Fix a typo in comments: bellow -> below.

Reported-by: Matthew Wilcox <willy@infradead.org>
Signed-off-by: Andrew Kreimer <algonell@gmail.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20241006130829.13967-1-algonell@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: phy: aquantia: allow forcing order of MDI pairs

Despite supporting Auto MDI-X, it looks like Aquantia only supports
swapping pair (1,2) with pair (3,6) like it used to be for MDI-X on
100MBit/s networks.

When all 4 pairs are in use (for 1000MBit/s or faster) the link does not
come up with pair order is not configured correctly, either using
MDI_CFG pin or using the "PMA Receive Reserved Vendor Provisioning 1"
register.

Normally, the order of MDI pairs being either ABCD or DCBA is configured
by pulling the MDI_CFG pin.

However, some hardware designs require overriding the value configured
by that bootstrap pin. The PHY allows doing that by setting a bit in
"PMA Receive Reserved Vendor Provisioning 1" register which allows
ignoring the state of the MDI_CFG pin and another bit configuring
whether the order of MDI pairs should be normal (ABCD) or reverse
(DCBA). Pair polarity is not affected and remains identical in both
settings.

Introduce property "marvell,mdi-cfg-order" which allows forcing either
normal or reverse order of the MDI pairs from DT.

If the property isn't present, the behavior is unchanged and MDI pair
order configuration is untouched (ie. either the result of MDI_CFG pin
pull-up/pull-down, or pair order override already configured by the
bootloader before Linux is started).

Forcing normal pair order is required on the Adtran SDG-8733A Wi-Fi 7
residential gateway.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/9ed760ff87d5fc456f31e407ead548bbb754497d.1728058550.git.daniel@makrotopia.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

dt-bindings: net: marvell,aquantia: add property to override MDI_CFG

Usually the MDI pair order reversal configuration is defined by
bootstrap pin MDI_CFG. Some designs, however, require overriding the MDI
pair order and force either normal or reverse order.

Add property 'marvell,mdi-cfg-order' to allow forcing either normal or
reverse order of the MDI pairs.

Signed-off-by: Daniel Golle <daniel@makrotopia.org>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://patch.msgid.link/7ccf25d6d7859f1ce9983c81a2051cfdfb0e0a99.1728058550.git.daniel@makrotopia.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'selftests-mlxsw-stabilize-red-tests'

Petr Machata says:

====================
selftests: mlxsw: Stabilize RED tests

Tweak the mlxsw-specific RED selftests to increase stability on
Spectrum-3 and Spectrum-4 machines.
====================

Link: https://patch.msgid.link/cover.1728316370.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

selftests: mlxsw: sch_red_core: Lower TBF rate

The RED test uses a pair of TBF shapers. The first to get predictably-sized
stream of traffic, and second to get a 100% saturated chokepoint. To this
chokepoint it injects individual packets. Because the chokepoint is
saturated, these additional packets go straight to the backlog. This allows
the test to check RED behavior across various queue sizes.

The shapers are rated at 1Gbps, for historical reasons (before mlxsw
supported TBF offload, the test used port speed to create the chokepoints).
Machines with a low-power CPU may have trouble consistently generating
1Gbps of traffic, and the test then spuriously fails.

Instead, drop the rate to 200Mbps (Spectrum has a guaranteed shaper rate
granularity of 200Mbps, so anything lower is not guaranteed to work well).
Because that means fewer packets will be mirrored in the ECN-mark test,
adjust the passing condition accordingly.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Link: https://patch.msgid.link/c6712f9c5de75ae0bc2ab3d8ea7d92aaaf93af95.1728316370.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

selftests: mlxsw: sch_red_core: Send more packets for drop tests

This test works by injecting into a port with a maxed-out queue a couple
packets and checks if a corresponding number of packets were dropped. This
has worked well on Spectrum<4, but on Spectrum-4 it has been noisy. This
is in line with the observation that on Spectrum-4, queue size tends to
fluctuate more. A handful of packets could then still be accepted to the
queue even though it was nominally full just recently.

In order to accommodate this behavior, send many more packets. The buffer
can fit N extra packets, but not N% packets. This therefore allows us to
set wider absolute margins, while actually narrowing them relatively.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Ido Schimmel <idosch@nvidia.com>
Link: https://patch.msgid.link/abc869b9f6003d400d6293ddd5edb2f4517f44d5.1728316370.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

selftests: mlxsw: sch_red_core: Sleep before querying queue depth

The qdisc stats are taken from the port's periodic HW stats, which are
updated once a second. We try to accommodate the latency by using busywait
in build_backlog().

The issue in that seems to be that when do_mark_test() builds the backlog,
it makes the decision whether to send more packets based on the first
instance of the queue depth stat exceeding the current value, when in fact
more traffic is on the way and the queue depth would increase further. This
leads to failures in TC 1 of mark-mirror test, where we see the following
failure:

TEST: TC 0: marked packets mirror'd                                 [ OK ]
TEST: TC 1: marked packets mirror'd                                 [FAIL]
        Spurious packets (1680 -> 2290) observed without buffer pressure

Fix by waiting for the full second before reading the queue depth for the
first time, to make sure it reflects all in-flight traffic.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Amit Cohen <amcohen@nvidia.com>
Link: https://patch.msgid.link/321dcf8b3e9a1f0766429c8cf3e3f1746f1bc375.1728316370.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

selftests: mlxsw: sch_red_core: Increase backlog size tolerance

Backlog fluctuates on Spectrum-4 much more than on <4. In practice we can
sample queue depth values going from about -12% to about +7% of the
configured RED limit. The test which checks the queue size has a limit of
+-10%, and as a result often fails. We attempted to fix the issue by
busywaiting for several seconds hoping to get within the bounds, but that
still proved to be too noisy (or the wait time would be impractically
long). Unfortunately we have to bump the value tolerance from 10% to 15%,
which in this patch do.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Amit Cohen <amcohen@nvidia.com>
Link: https://patch.msgid.link/f54950df2a8fcba46c3ddc1053376352fa2e592b.1728316370.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

selftests: mlxsw: sch_red_ets: Increase required backlog

Backlog fluctuates on Spectrum-4 much more than on <4. Increasing the
desired backlog seems to help, as the constant fluctuations do not overlap
into the territory where packets are marked.

Signed-off-by: Petr Machata <petrm@nvidia.com>
Reviewed-by: Amit Cohen <amcohen@nvidia.com>
Link: https://patch.msgid.link/0821fb3aa8bb6a6c0d3000baab04995517c9a0cc.1728316370.git.petrm@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: phy: smsc: use devm_clk_get_optional_enabled_with_rate()

Fold the separate call to clk_set_rate() into the clock getter.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20241007134100.107921-1-brgl@bgdev.pl
Signed-off-by: Jakub Kicinski <kuba@kernel.org>