eth: fbnic: Add message parsing for FW messages

Add FW message formatting and parsing. The TLV format should
look very familiar to those familiar with netlink.
Since we don't have to deal with backward compatibility
we tweaked the format a little to make it easier to deal
with, and more appropriate for tightly coupled interfaces
like driver<>FW communication.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://patch.msgid.link/172079936754.1778861.1029830244010564007.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

eth: fbnic: Add register init to set PCIe/Ethernet device config

As a part of enabling the device the first step is to configure the AXI and
Ethernet interfaces to allow for basic traffic. This consists of
configuring several registers related to the PCIe and Ethernet FIFOs as
well as configuring the handlers for moving traffic between entities.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://patch.msgid.link/172079936376.1778861.15942501417449077552.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

eth: fbnic: Allocate core device specific structures and devlink interface

At the core of the fbnic device will be the devlink interface. This
interface will eventually provide basic functionality in the event that
there are any issues with the network interface.

Add support for allocating the MSI-X vectors and setting up the BAR
mapping. With this we can start enabling various subsystems and start
brining up additional interfaces such the AXI fabric and the firmware
mailbox.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Link: https://patch.msgid.link/172079936012.1778861.4670986685222676467.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

eth: fbnic: Add scaffolding for Meta's NIC driver

Create a bare-bones PCI driver for Meta's NIC.
Subsequent changes will flesh it out.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/172079935646.1778861.9710282776096050607.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

PCI: Add Meta Platforms vendor ID

Add Meta as a vendor ID for PCI devices so we can use the macro for future
drivers.

Signed-off-by: Alexander Duyck <alexanderduyck@fb.com>
Acked-by: Bjorn Helgaas <bhelgaas@google.com>
Link: https://patch.msgid.link/172079935272.1778861.13619056509276833225.stgit@ahduyck-xeon-server.home.arpa
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge tag 'aux-sysfs-irqs' of git://git./linux/kernel/git/mellanox/linux

Saeed Mahameed says:

====================
aux-sysfs-irqs

Shay Says:
==========
Introduce auxiliary bus IRQs sysfs

Today, PCI PFs and VFs, which are anchored on the PCI bus, display their
IRQ information in the <pci_device>/msi_irqs/<irq_num> sysfs files.  PCI
subfunctions (SFs) are similar to PFs and VFs and these SFs are anchored
on the auxiliary bus. However, these PCI SFs lack such IRQ information
on the auxiliary bus, leaving users without visibility into which IRQs
are used by the SFs. This absence makes it impossible to debug
situations and to understand the source of interrupts/SFs for
performance tuning and debug.

Additionally, the SFs are multifunctional devices supporting RDMA,
network devices, clocks, and more, similar to their peer PCI PFs and
VFs. Therefore, it is desirable to have SFs' IRQ information available
at the bus/device level.

To overcome the above limitations, this short series extends the
auxiliary bus to display IRQ information in sysfs, similar to that of
PFs and VFs.

It adds an 'irqs' directory under the auxiliary device and includes an
<irq_num> sysfs file within it.

For example:
$ ls /sys/bus/auxiliary/devices/mlx5_core.sf.1/irqs/
50  51  52  53  54  55  56  57  58

Patch summary:
patch-1 adds auxiliary bus to support irqs used by auxiliary device
patch-2 mlx5 driver using exposing irqs for PCI SF devices via auxiliary
        bus
==========

* tag 'aux-sysfs-irqs' of git://git.kernel.org/pub/scm/linux/kernel/git/mellanox/linux:
  net/mlx5: Expose SFs IRQs
  driver core: auxiliary bus: show auxiliary device IRQs
  RDMA/mlx5: Add Qcounters req_transport_retries_exceeded/req_rnr_retries_exceeded
  net/mlx5: Reimplement write combining test
====================

Link: https://patch.msgid.link/20240711213140.256997-1-saeed@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'flower-rework-tca_flower_key_enc_flags-usage'

Asbjørn Sloth Tønnesen says:

====================
flower: rework TCA_FLOWER_KEY_ENC_FLAGS usage

This series reworks the recently added TCA_FLOWER_KEY_ENC_FLAGS
attribute, to be more like TCA_FLOWER_KEY_FLAGS, and use the unused
u32 flags field in FLOW_DISSECTOR_KEY_ENC_CONTROL, instead of adding
a new flags field as FLOW_DISSECTOR_KEY_ENC_FLAGS.

I have defined the new FLOW_DIS_F_* and TCA_FLOWER_KEY_FLAGS_*
flags to co-exist with the existing flags, so the meaning
of the flags field in struct flow_dissector_key_control is not
depending on the context it is used in. If we run out of bits
then we can always split them up later, if we really want to.
Future flags might also be valid in both contexts.

iproute2 RFC v2 patch:
https://lore.kernel.org/560bcd549ca8ab24b1ad5abe352580a621f6d426.1720790774.git.dcaratti@redhat.com/

v3: https://lore.kernel.org/20240709163825.1210046-1-ast@fiberby.net/
v2: https://lore.kernel.org/20240705133348.728901-1-ast@fiberby.net/
v1: https://lore.kernel.org/20240703104600.455125-1-ast@fiberby.net/
RFC: https://lore.kernel.org/20240611235355.177667-1-ast@fiberby.net/
====================

Link: https://patch.msgid.link/20240713021911.1631517-1-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/sched: cls_flower: propagate tca[TCA_OPTIONS] to NL_REQ_ATTR_CHECK

NL_REQ_ATTR_CHECK() is used in fl_set_key_flags() to set
extended attributes about the origin of an error, this
patch propagates tca[TCA_OPTIONS] through.

Before this patch:

$ sudo ./tools/net/ynl/cli.py --spec Documentation/netlink/specs/tc.yaml \
 --do newtfilter --json '{
"chain": 0, "family": 0, "handle": 4, "ifindex": 22,
"info": 262152,
"kind": "flower",
"options": {
"flags": 0, "key-enc-flags": 8,
"key-eth-type": 2048 },
"parent": 4294967283 }'
Netlink error: Invalid argument
nl_len = 68 (52) nl_flags = 0x300 nl_type = 2
        error: -22
        extack: {'msg': 'Missing flags mask',
                 'miss-type': 111}

After this patch:

[same cmd]
Netlink error: Invalid argument
nl_len = 76 (60) nl_flags = 0x300 nl_type = 2
        error: -22
        extack: {'msg': 'Missing flags mask',
                 'miss-type': 111, 'miss-nest': 56}

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Link: https://patch.msgid.link/20240713021911.1631517-14-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

flow_dissector: set encapsulation control flags for non-IP

Make sure to set encapsulated control flags also for non-IP
packets, such that it's possible to allow matching on e.g.
TUNNEL_OAM on a geneve packet carrying a non-IP packet.

Suggested-by: Davide Caratti <dcaratti@redhat.com>
Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-13-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

flow_dissector: cleanup FLOW_DISSECTOR_KEY_ENC_FLAGS

Now that TCA_FLOWER_KEY_ENC_FLAGS is unused, as it's
former data is stored behind TCA_FLOWER_KEY_ENC_CONTROL,
then remove the last bits of FLOW_DISSECTOR_KEY_ENC_FLAGS.

FLOW_DISSECTOR_KEY_ENC_FLAGS is unreleased, and have been
in net-next since 2024-06-04.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-12-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

doc: netlink: specs: tc: flower: add enc-flags

Describe key-enc-flags and key-enc-flags-mask.

These are defined similarly to key-flags and key-flags-mask.

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20240713021911.1631517-11-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/sched: cls_flower: rework TCA_FLOWER_KEY_ENC_FLAGS usage

This patch changes how TCA_FLOWER_KEY_ENC_FLAGS is used, so that
it is used with TCA_FLOWER_KEY_FLAGS_* flags, in the same way as
TCA_FLOWER_KEY_FLAGS is currently used.

Where TCA_FLOWER_KEY_FLAGS uses {key,mask}->control.flags, then
TCA_FLOWER_KEY_ENC_FLAGS now uses {key,mask}->enc_control.flags,
therefore {key,mask}->enc_flags is now unused.

As the generic fl_set_key_flags/fl_dump_key_flags() is used with
encap set to true, then fl_{set,dump}_key_enc_flags() is removed.

This breaks unreleased userspace API (net-next since 2024-06-04).

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-10-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/sched: cls_flower: add tunnel flags to fl_{set,dump}_key_flags()

Prepare to set and dump the tunnel flags.

This code won't see any of these flags yet, as these flags
aren't allowed by the NLA_POLICY_MASK, and the functions
doesn't get called with encap set to true yet.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-9-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

flow_dissector: set encapsulated control flags from tun_flags

Set the new FLOW_DIS_F_TUNNEL_* encapsulated control flags, based
on if their counter-part is set in tun_flags.

These flags are not userspace visible yet, as the code to dump
encapsulated control flags will first be added, and later activated
in the following patches.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-8-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

flow_dissector: prepare for encapsulated control flags

Rename skb_flow_dissect_set_enc_addr_type() to
skb_flow_dissect_set_enc_control(), and make it set both
addr_type and flags in FLOW_DISSECTOR_KEY_ENC_CONTROL.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-7-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/sched: cls_flower: add policy for TCA_FLOWER_KEY_FLAGS

This policy guards fl_set_key_flags() from seeing flags
not used in the context of TCA_FLOWER_KEY_FLAGS.

In order For the policy check to be performed with the
correct endianness, then we also needs to change the
attribute type to NLA_BE32 (Thanks Davide).

TCA_FLOWER_KEY_FLAGS{,_MASK} already has a be32 comment
in include/uapi/linux/pkt_cls.h.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-6-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/sched: cls_flower: prepare fl_{set,dump}_key_flags() for ENC_FLAGS

Prepare fl_set_key_flags/fl_dump_key_flags() for use with
TCA_FLOWER_KEY_ENC_FLAGS{,_MASK}.

This patch adds an encap argument, similar to fl_set_key_ip/
fl_dump_key_ip(), and determine the flower keys based on the
encap argument, and use them in the rest of the two functions.

Since these functions are so far, only called with encap set false,
then there is no functional change.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Tested-by: Davide Caratti <dcaratti@redhat.com>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-5-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/sched: flower: define new tunnel flags

Define new TCA_FLOWER_KEY_FLAGS_* flags for use in struct
flow_dissector_key_control, covering the same flags as
currently exposed through TCA_FLOWER_KEY_ENC_FLAGS.

Put the new flags under FLOW_DIS_F_*. The idea is that we can
later, move the existing flags under FLOW_DIS_F_* as well.

The ynl flag names have been taken from the RFC iproute2 patch.

Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20240713021911.1631517-4-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

doc: netlink: specs: tc: describe flower control flags

Describe the flower control flags, and use them
for key-flags and key-flags-mask.

The flag names have been taken from iproute2.

Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Reviewed-by: Donald Hunter <donald.hunter@gmail.com>
Link: https://patch.msgid.link/20240713021911.1631517-3-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net/sched: flower: refactor control flag definitions

Redefine the flower control flags as an enum, so they are
included in BTF info.

Make the kernel-side enum a more explicit superset of
TCA_FLOWER_KEY_FLAGS_*, new flags still need to be added to
both enums, but at least the bit position only has to be
defined once.

FLOW_DIS_ENCAPSULATION is never set for mask, so it can't be
exposed to userspace in an unsupported flags mask error message,
so it will be placed one bit position above the last uAPI flag.

Suggested-by: Alexander Lobakin <aleksander.lobakin@intel.com>
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Asbjørn Sloth Tønnesen <ast@fiberby.net>
Reviewed-by: Davide Caratti <dcaratti@redhat.com>
Link: https://patch.msgid.link/20240713021911.1631517-2-ast@fiberby.net
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Revert "net: mvpp2: Improve data types and use min()"

This reverts commit f7023b3d697c6a7dfe2d9c70e0d8c2c580ccbd76.

Russell indicates that assuming 32bits are sufficient isn't
necessarily safe.

Link: https://lore.kernel.org/20240711154741.174745-1-thorsten.blum@toblux.com/
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

llc: Constify struct llc_sap_state_trans

'struct llc_sap_state_trans' are not modified in this driver.

Constifying this structure moves some data to a read-only section, so
increase overall security.

On a x86_64, with allmodconfig, as an example:
Before:
======
   text    data     bss     dec     hex filename
    339     456      24     819     333 net/llc/llc_s_st.o

After:
=====
   text    data     bss     dec     hex filename
    683     144       0     827     33b net/llc/llc_s_st.o

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/9d17587639195ee94b74ff06a11ef97d1833ee52.1720973710.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

llc: Constify struct llc_conn_state_trans

'struct llc_conn_state_trans' are not modified in this driver.

Constifying this structure moves some data to a read-only section, so
increase overall security.

On a x86_64, with allmodconfig, as an example:
Before:
======
   text    data     bss     dec     hex filename
  13923   10896      32   24851    6113 net/llc/llc_c_st.o

After:
=====
   text    data     bss     dec     hex filename
  21859    3328       0   25187    6263 net/llc/llc_c_st.o

Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/87cda89e4c9414e71d1a54bb1eb491b0e7f70375.1720973029.git.christophe.jaillet@wanadoo.fr
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

dsa: lan9303: consistent naming for PHY address parameter

Name it 'addr' instead of 'port' or 'phy'.

Signed-off-by: Christian Eggers <ceggers@arri.de>
Link: https://patch.msgid.link/20240715123050.21202-1-ceggers@arri.de
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge tag 'for-net-next-2024-07-15' of git://git./linux/kernel/git/bluetooth/bluetooth-next

Luiz Augusto von Dentz says:

====================
bluetooth-next pull request for net-next:

 - qca: use the power sequencer for QCA6390
 - btusb: mediatek: add ISO data transmission functions
 - hci_bcm4377: Add BCM4388 support
 - btintel: Add support for BlazarU core
 - btintel: Add support for Whale Peak2
 - btnxpuart: Add support for AW693 A1 chipset
 - btnxpuart: Add support for IW615 chipset
 - btusb: Add Realtek RTL8852BE support ID 0x13d3:0x3591

* tag 'for-net-next-2024-07-15' of git://git.kernel.org/pub/scm/linux/kernel/git/bluetooth/bluetooth-next: (71 commits)
  Bluetooth: btmtk: Mark all stub functions as inline
  Bluetooth: hci_qca: Fix build error
  Bluetooth: hci_qca: use the power sequencer for wcn7850 and wcn6855
  Bluetooth: hci_qca: make pwrseq calls the default if available
  Bluetooth: hci_qca: unduplicate calls to hci_uart_register_device()
  Bluetooth: hci_qca: schedule a devm action for disabling the clock
  dt-bindings: bluetooth: qualcomm: describe the inputs from PMU for wcn7850
  Bluetooth: btnxpuart: Fix warnings for suspend and resume functions
  Bluetooth: btnxpuart: Add system suspend and resume handlers
  Bluetooth: btnxpuart: Add support for IW615 chipset
  Bluetooth: btnxpuart: Add support for AW693 A1 chipset
  Bluetooth: btintel: Add support for Whale Peak2
  Bluetooth: btintel: Add support for BlazarU core
  Bluetooth: btusb: mediatek: add ISO data transmission functions
  Bluetooth: btmtk: move btusb_recv_acl_mtk to btmtk.c
  Bluetooth: btmtk: move btusb_mtk_[setup, shutdown] to btmtk.c
  Bluetooth: btmtk: move btusb_mtk_hci_wmt_sync to btmtk.c
  Bluetooth: btusb: add callback function in btusb suspend/resume
  Bluetooth: btmtk: rename btmediatek_data
  Bluetooth: btusb: mediatek: return error for failed reg access
  ...
====================

Link: https://patch.msgid.link/20240715142543.303944-1-luiz.dentz@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'net-make-timestamping-selectable'

First part of "net: Make timestamping selectable" from Kory Maincent.
Change the driver-facing type already to lower rebasing pain.

Link: https://lore.kernel.org/20240709-feature_ptp_netnext-v17-0-b5317f50df2a@bootlin.com/
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: Add struct kernel_ethtool_ts_info

In prevision to add new UAPI for hwtstamp we will be limited to the struct
ethtool_ts_info that is currently passed in fixed binary format through the
ETHTOOL_GET_TS_INFO ethtool ioctl. It would be good if new kernel code
already started operating on an extensible kernel variant of that
structure, similar in concept to struct kernel_hwtstamp_config vs struct
hwtstamp_config.

Since struct ethtool_ts_info is in include/uapi/linux/ethtool.h, here
we introduce the kernel-only structure in include/linux/ethtool.h.
The manual copy is then made in the function called by ETHTOOL_GET_TS_INFO.

Acked-by: Shannon Nelson <shannon.nelson@amd.com>
Acked-by: Alexandra Winter <wintera@linux.ibm.com>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20240709-feature_ptp_netnext-v17-6-b5317f50df2a@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: net_tstamp: Add unspec field to hwtstamp_source enumeration

Prepare for future support of saving hwtstamp source in PTP xarray by
introducing HWTSTAMP_SOURCE_UNSPEC to hwtstamp_source enum, setting it
to 0 to match old behavior of no source defined.

Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20240709-feature_ptp_netnext-v17-5-b5317f50df2a@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: Change the API of PHY default timestamp to MAC

Change the API to select MAC default time stamping instead of the PHY.
Indeed the PHY is closer to the wire therefore theoretically it has less
delay than the MAC timestamping but the reality is different. Due to lower
time stamping clock frequency, latency in the MDIO bus and no PHC hardware
synchronization between different PHY, the PHY PTP is often less precise
than the MAC. The exception is for PHY designed specially for PTP case but
these devices are not very widespread. For not breaking the compatibility
default_timestamp flag has been introduced in phy_device that is set by
the phy driver to know we are using the old API behavior.

Reviewed-by: Rahul Rameshbabu <rrameshbabu@nvidia.com>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20240709-feature_ptp_netnext-v17-4-b5317f50df2a@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Bluetooth: btmtk: Mark all stub functions as inline

Several recent patches added static stubs to btmtk.h without the inline
keyword, which causes instances of -Wunused-function when those stubs
are not used anywhere in a file that includes the header:

  In file included from drivers/bluetooth/btusb.c:28:
  drivers/bluetooth/btmtk.h:254:13: warning: 'btmtk_fw_get_filename' defined but not used [-Wunused-function]
    254 | static void btmtk_fw_get_filename(char *buf, size_t size, u32 dev_id,
        |             ^~~~~~~~~~~~~~~~~~~~~
  drivers/bluetooth/btmtk.h:249:12: warning: 'btmtk_process_coredump' defined but not used [-Wunused-function]
    249 | static int btmtk_process_coredump(struct hci_dev *hdev, struct sk_buff *skb)
        |            ^~~~~~~~~~~~~~~~~~~~~~
  drivers/bluetooth/btmtk.h:243:12: warning: 'btmtk_register_coredump' defined but not used [-Wunused-function]
    243 | static int btmtk_register_coredump(struct hci_dev *hdev, const char *name,
        |            ^~~~~~~~~~~~~~~~~~~~~~~
  drivers/bluetooth/btmtk.h:233:12: warning: 'btmtk_setup_firmware' defined but not used [-Wunused-function]
    233 | static int btmtk_setup_firmware(struct hci_dev *hdev, const char *fwname,
        |            ^~~~~~~~~~~~~~~~~~~~
  drivers/bluetooth/btmtk.h:227:12: warning: 'btmtk_setup_firmware_79xx' defined but not used [-Wunused-function]
    227 | static int btmtk_setup_firmware_79xx(struct hci_dev *hdev, const char *fwname,
        |            ^~~~~~~~~~~~~~~~~~~~~~~~~

Add inline to all the stubs in btmtk.h (even ones that do not currently
have any warnings associated with them) to ensure there are never unused
function warnings from these stubs, as is customary for the kernel.

Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: hci_qca: Fix build error

This fixes the following build error introduced by b80f4e3b8cb8
("Bluetooth: hci_qca: schedule a devm action for disabling the clock"):

drivers/bluetooth/hci_qca.c: In function ‘qca_serdev_remove’:
drivers/bluetooth/hci_qca.c:2501:2: error: label at end of compound statement
 2501 |  default:
      |  ^~~~~~~

Fixes: b80f4e3b8cb8 ("Bluetooth: hci_qca: schedule a devm action for disabling the clock")
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: hci_qca: use the power sequencer for wcn7850 and wcn6855

In preparation for using the power sequencing subsystem on sm8650 boards
and X13s laptop let's make pwrseq the default for wcn7850 and wcn6855.

Both these models require an enable GPIO so we can safely assume that if
the property is not there, then we should try to get the power
sequencer. Due to how the pwrseq lookup works - checking the provider at
run-time - we cannot really do it the other way around as we'd get stuck
forever on -EPROBE_DEFER.

If the relevant OF node does have the 'enable-gpios' property, we
fallback to the existing code for backward compatibility with older DTs.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: hci_qca: make pwrseq calls the default if available

If the device has a power sequencing handle, use it first. Otherwise
fall back to whatever code already exists.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: hci_qca: unduplicate calls to hci_uart_register_device()

Now that all three branches of the switch end up doing the same thing,
we can move the call to hci_uart_register_device() past it and unify the
error message.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: hci_qca: schedule a devm action for disabling the clock

In preparation for unduplicating the hci_uart registration code,
schedule a devres action for disabling the SUSCLK rather than doing it
manually.

We cannot really use devm_clk_get_enabled() as we also set the rate
before enabling the clock. While this should in theory work, I don't
want to risk breaking existing users. One solution for the future is to
add devm_clk_get_enabled_with_rate() to the clock framework.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

dt-bindings: bluetooth: qualcomm: describe the inputs from PMU for wcn7850

Drop the inputs from the host and instead expect the Bluetooth node to
consume the outputs of the internal PMU. This model is closer to reality
than how we represent it now.

Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Acked-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: btnxpuart: Fix warnings for suspend and resume functions

This fixes the warnings 'nxp_serdev_resume' and 'nxp_serdev_suspend'
defined but not used.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Reported-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202407100518.06Xxq5PK-lkp@intel.com/
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: btnxpuart: Add system suspend and resume handlers

This adds handling for system suspend and resume. While the host enters
suspend state, the driver will drive the chip into low power state.

Similarly when system is resuming, the driver will wake up the chip.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: btnxpuart: Add support for IW615 chipset

This adds support for IW615 chipset with it's bootloader signature
and firmware file.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

net_tstamp: Add TIMESTAMPING SOFTWARE and HARDWARE mask

Timestamping software or hardware flags are often used as a group,
therefore adding these masks will ease future use.

I did not use SOF_TIMESTAMPING_SYS_HARDWARE flag as it is deprecated and
not used at all.

Reviewed-by: Willem de Bruijn <willemb@google.com>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Signed-off-by: Kory Maincent <kory.maincent@bootlin.com>
Link: https://patch.msgid.link/20240709-feature_ptp_netnext-v17-1-b5317f50df2a@bootlin.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Bluetooth: btnxpuart: Add support for AW693 A1 chipset

This adds support for AW693 A1 chipset with it's bootloader signature
and firmware file.

No firmware name change needed for AW693 A0 chipset.

Signed-off-by: Neeraj Sanjay Kale <neeraj.sanjaykale@nxp.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: btintel: Add support for Whale Peak2

from /sys/kernel/debug/usb/devices:

T:  Bus=03 Lev=01 Prnt=01 Port=09 Cnt=02 Dev#=  3 Spd=12   MxCh= 0
D:  Ver= 2.00 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=8087 ProdID=0039 Rev= 0.00
C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=100mA
I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=81(I) Atr=03(Int.) MxPS=  64 Ivl=1ms
E:  Ad=02(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=82(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   0 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   0 Ivl=1ms
I:  If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   9 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   9 Ivl=1ms
I:  If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  17 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  17 Ivl=1ms
I:  If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  25 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  25 Ivl=1ms
I:  If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  33 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  33 Ivl=1ms
I:  If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  49 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  49 Ivl=1ms
I:  If#= 1 Alt= 6 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  63 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  63 Ivl=1ms

Signed-off-by: Kiran K <kiran.k@intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: btintel: Add support for BlazarU core

Add support for BlazarU core (CNVi).

Signed-off-by: Kiran K <kiran.k@intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: btusb: mediatek: add ISO data transmission functions

This patch implements functions for ISO data send and receive in btusb
driver for MediaTek's controller.

MediaTek defines a specific interrupt endpoint for ISO data transmissin
because the characteristics of interrupt endpoint are similar to the
application of ISO data which can support guaranteed transmissin
bandwidth, enough maximum data length and error checking mechanism.

Driver sets up ISO interface and endpoints in btusb_mtk_setup and clears
the setup in btusb_mtk_shutdown. These flow can't move to btmtk.c due to
btusb_driver is only defined in btusb.c when claiming/relaesing interface.
ISO packet anchor stops when driver suspending and resubmit interrupt urb
for ISO data when driver resuming.

Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: btmtk: move btusb_recv_acl_mtk to btmtk.c

Move btusb_recv_acl_mtk from btusb.c to btmtk.c which holds
vendor specific stuff and would make btusb.c clean.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: btmtk: move btusb_mtk_[setup, shutdown] to btmtk.c

Move btusb_mtk_[setup, shutdown] and related function from
btusb.c to btmtk.c which holds vendor specific stuff and
would make btusb.c clean.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: btmtk: move btusb_mtk_hci_wmt_sync to btmtk.c

Move btusb_mtk_hci_wmt_sync from btusb.c to btmtk.c which holds
vendor specific stuff and would make btusb.c clean.

Add usb.h header to btmtksdio.c/btmtkuart.c for usb related element
defined in btmtk.h

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: btusb: add callback function in btusb suspend/resume

Add suspend/resum callback function in btusb_data which are reserved
for vendor specific usage during suspend/resume. hdev->suspend will be
added before stop traffic in btusb_suspend and hdev-> resume will be
added after resubmit urb in btusb_resume.

Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: btmtk: rename btmediatek_data

Rename btmediatek_data to have a consistent prefix throughout the driver.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: btusb: mediatek: return error for failed reg access

Update the code to immediately return an error code if accessing a
related register fails. This ensures that our desired logic for
subsequent register operations is maintained and allows us to promptly
catch any unexpected errors.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: btusb: mediatek: remove the unnecessary goto tag

Remove the unnecessary goto tag whether there is an error or not, we have
to free the buffer at the end of the function.

Signed-off-by: Sean Wang <sean.wang@mediatek.com>
Signed-off-by: Chris Lu <chris.lu@mediatek.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: Remove hci_request.{c,h}

This removes hci_request.{c,h} since it shall no longer be used.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: hci_sync: Remove remaining dependencies of hci_request

This removes the dependencies of hci_req_init and hci_request_cancel_all
from hci_sync.c.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: hci_sync: Move handling of interleave_scan

This moves handling of interleave_scan work to hci_sync.c since
hci_request.c is deprecated.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: hci_core: Don't use hci_prepare_cmd

This replaces the instance of hci_prepare_cmd with hci_cmd_sync_alloc
since the former is part of hci_request.c which is considered
deprecated.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: hci_core: Remove usage of hci_req_sync

hci_request functions are considered deprecated so this replaces the
usage of hci_req_sync with hci_inquiry_sync.

Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

dt-bindings: net: bluetooth: convert MT7622 Bluetooth to the json-schema

This helps validating DTS files. Introduced changes:
1. Dropped serial details from example
2. Added required example include

Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: Fix usage of __hci_cmd_sync_status

__hci_cmd_sync_status shall only be used if hci_req_sync_lock is _not_
required which is not the case of hci_dev_cmd so it needs to use
hci_cmd_sync_status which uses hci_req_sync_lock internally.

Fixes: f1a8f402f13f ("Bluetooth: L2CAP: Fix deadlock")
Reported-by: Pauli Virtanen <pav@iki.fi>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: btusb: Add Realtek RTL8852BE support ID 0x13d3:0x3591

Add the support ID(0x13d3, 0x3591) to usb_device_id table for
Realtek RTL8852BE.

The device table is as follows:

T:  Bus=01 Lev=02 Prnt=03 Port=00 Cnt=01 Dev#=  5 Spd=12   MxCh= 0
D:  Ver= 1.00 Cls=e0(wlcon) Sub=01 Prot=01 MxPS=64 #Cfgs=  1
P:  Vendor=13d3 ProdID=3591 Rev= 0.00
S:  Manufacturer=Realtek
S:  Product=Bluetooth Radio
S:  SerialNumber=00e04c000001
C:* #Ifs= 2 Cfg#= 1 Atr=e0 MxPwr=500mA
I:* If#= 0 Alt= 0 #EPs= 3 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=81(I) Atr=03(Int.) MxPS=  16 Ivl=1ms
E:  Ad=02(O) Atr=02(Bulk) MxPS=  64 Ivl=0ms
E:  Ad=82(I) Atr=02(Bulk) MxPS=  64 Ivl=0ms
I:* If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   0 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   0 Ivl=1ms
I:  If#= 1 Alt= 1 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=   9 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=   9 Ivl=1ms
I:  If#= 1 Alt= 2 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  17 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  17 Ivl=1ms
I:  If#= 1 Alt= 3 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  25 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  25 Ivl=1ms
I:  If#= 1 Alt= 4 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  33 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  33 Ivl=1ms
I:  If#= 1 Alt= 5 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
E:  Ad=03(O) Atr=01(Isoc) MxPS=  49 Ivl=1ms
E:  Ad=83(I) Atr=01(Isoc) MxPS=  49 Ivl=1ms

Cc: stable@vger.kernel.org
Signed-off-by: Wentao Guan <guanwentao@uniontech.com>
Signed-off-by: WangYuli <wangyuli@uniontech.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: hci_core: cleanup struct hci_dev

Remove unused and set but otherwise unused 'discovery_old_state'
and 'sco_last_tx' members of 'struct hci_dev'. The first one is
a leftover after commit 182ee45da083 ("Bluetooth: hci_sync: Rework
hci_suspend_notifier"); the second one is originated from ancient
2.4.19 and I was unable to find any actual use since that.

Signed-off-by: Dmitry Antipov <dmantipov@yandex.ru>
Reviewed-by: Paul Menzel <pmenzel@molgen.mpg.de>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: btintel: Fix the sfi name for BlazarU

Use INTEL_HW_VARIANT() instead of CNVi Id to decide to load Intermediate
Loader (IML) image. Fix the driver loading incorrect firmware for
BlazarU product.

dmesg:
.....
[146.111834] Bluetooth: hci0: Minimum firmware build 1 week 10 2014
[146.111839] Bluetooth: hci0: Bootloader timestamp 2022.18 buildtype 1 build 16362
[146.111848] Bluetooth: hci0: No support for _PRR ACPI method
[146.112204] Bluetooth: hci0: Failed to load Intel firmware file intel/ibt-0291-0291-iml.sfi (-2)

Fixes: 164c62f958f8 ("Bluetooth: btintel: Add firmware ID to firmware name")
Reported-by: Tsigan, Vladislav <vladislav.tsigan@intel.com>
Signed-off-by: Kiran K <kiran.k@intel.com>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: qca: don't disable power management for QCA6390

We unnecessarily fallthrough the case for QCA6390 when initializing the
device and hit the condition where - due to the lack of the enable-gpio
- we disable power management despite using the power sequencer. We don't
need to look for clocks on this model so it makes more sense to just
register the hci device and break the switch.

Reported-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Fixes: 9a15ce685706 ("Bluetooth: qca: use the power sequencer for QCA6390")
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Acked-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Tested-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> # RB5
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: hci: fix build when POWER_SEQUENCING=m

BT_HCIUART_QCA, which is a bool and an HCI UART component, uses now
power sequencing, so we need to be sure BT_HCIUART_QCA=y and
POWER_SEQUENCING=m cannot happen:

  /usr/bin/aarch64-linux-gnu-ld: drivers/bluetooth/hci_qca.o: in function `qca_serdev_probe':
  drivers/bluetooth/hci_qca.c:2401:(.text+0xc48): undefined reference to `devm_pwrseq_get'
  /usr/bin/aarch64-linux-gnu-ld: drivers/bluetooth/hci_qca.o: in function `qca_power_shutdown':
  drivers/bluetooth/hci_qca.c:2192:(.text+0x29e8): undefined reference to `pwrseq_power_off'
  /usr/bin/aarch64-linux-gnu-ld: drivers/bluetooth/hci_qca.o: in function `qca_power_on':
  drivers/bluetooth/hci_qca.c:1808:(.text+0x3390): undefined reference to `pwrseq_power_on'

Fixes: 9a15ce685706 ("Bluetooth: qca: use the power sequencer for QCA6390")
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: btintel: Fix spelling of *intermediate* in comment

Fixes: f3b845e0aea3 ("Bluetooth: btintel: Add support to download intermediate loader")
Cc: Kiran K <kiran.k@intel.com>
Signed-off-by: Paul Menzel <pmenzel@molgen.mpg.de>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Bluetooth: qca: use the power sequencer for QCA6390

Use the pwrseq subsystem's consumer API to run the power-up sequence for
the Bluetooth module of the QCA6390 package.

Tested-by: Amit Pundir <amit.pundir@linaro.org>
Tested-by: Neil Armstrong <neil.armstrong@linaro.org> # on SM8550-QRD, SM8650-QRD & SM8650-HDK
Tested-by: Caleb Connolly <caleb.connolly@linaro.org> # OnePlus 8T
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

dt-bindings: net: bluetooth: qualcomm: describe regulators for QCA6390

QCA6390 has a compatible listed in the bindings but is missing the
regulators description. Add the missing supply property and list the
required ones in the allOf section.

Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Bartosz Golaszewski <bartosz.golaszewski@linaro.org>
Signed-off-by: Luiz Augusto von Dentz <luiz.von.dentz@intel.com>

Merge tag 'pwrseq-initial-for-v6.11' of git://git./linux/kernel/git/brgl/linux into HEAD

Initial implementation of the power sequencing subsystem for linux v6.11

net: ethernet: lantiq_etop: remove redundant device name setup

The same name is set when allocating the netdevice structure in the
alloc_etherdev_mq()->alloc_etherrdev_mqs() function. Therefore, there
is no need to manually set it.

This fixes CheckPatch warnings:
WARNING: Prefer strscpy over strcpy - see: https://github.com/KSPP/linux/issues/88
strcpy(dev->name, "eth%d");

Signed-off-by: Aleksander Jan Bajkowski <olek2@wp.pl>
Link: https://patch.msgid.link/20240713170920.863171-1-olek2@wp.pl
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'net-dsa-vsc73xx-implement-vlan-operations'

Pawel Dembicki says:

====================
net: dsa: vsc73xx: Implement VLAN operations

This patch series is a result of splitting a larger patch series [0],
where some parts was merged before.

The first patch implements port state configuration, which is required
for bridge functionality. STP frames are not forwarded at this moment.
BPDU frames are only forwarded from/to the PI/SI interface.
For more information, see chapter 2.7.1 (CPU Forwarding) in the
datasheet.

Patches 2, 7-9 and 11 provide a basic implementation of tag_8021q
functionality with QinQ support, without VLAN filtering in
the bridge and simple VLAN awareness in VLAN filtering mode.

Patches 3-6 came from Vladimir Oltean. They prepare for making
tag8021q more common. VSC73XX uses very similar tag recognition,
and some code from tag_sja1105 could be moved to tag_8021q for
common use.

Patch 10 is preparation for use tag_8021q bridge functions as generic
implementation of the 'ds->ops->port_bridge_*()'.

Patch 12 is required to avoid problem with learning on standalone ports.

[0] https://patchwork.kernel.org/project/netdevbpf/list/?series=841034&state=%2A&archive=both
====================

Link: https://patch.msgid.link/20240713211620.1125910-1-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: dsa: vsc73xx: start treating the BR_LEARNING flag

This patch implements .port_pre_bridge_flags() and .port_bridge_flags(),
which are required for properly treating the BR_LEARNING flag. Also,
.port_stp_state_set() is tweaked and now disables learning for standalone
ports.

Disabling learning for standalone ports is required to avoid situations
where one port sees traffic originating from another, which could cause
packet drops.

Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-13-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: dsa: vsc73xx: Add bridge support

This patch adds bridge support for the vsc73xx driver.

The vsc73xx requires minimal operations and ithe generic
dsa_tag_8021q_bridge_* API is sufficient.
The forwarding matrix is managed by vsc73xx_port_stp_state_set() ->
vsc73xx_refresh_fwd_map()i routine, which is called immediately after
.port_bridge_join() and .port_bridge_leave().

Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-12-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: dsa: prepare 'dsa_tag_8021q_bridge_join' for standalone use

The 'dsa_tag_8021q_bridge_join' could be used as a generic implementation
of the 'ds->ops->port_bridge_join()' function. However, it is necessary
to synchronize their arguments.

This patch also moves the 'tx_fwd_offload' flag configuration line into
'dsa_tag_8021q_bridge_join' body. Currently, every (sja1105) driver sets
it, and the future vsc73xx implementation will also need it for
simplification.

Suggested-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-11-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: dsa: Define max num of bridges in tag8021q implementation

Max number of bridges in tag8021q implementation is strictly limited
by VBID size: 3 bits. But zero is reserved and only 7 values can be used.

This patch adds define which describe maximum possible value.

Suggested-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-10-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: dsa: vsc73xx: Implement the tag_8021q VLAN operations

This patch is a simple implementation of 802.1q tagging in the vsc73xx
driver. Currently, devices with DSA_TAG_PROTO_NONE are not functional.
The VSC73XX family doesn't provide any tag support for external Ethernet
ports.

The only option available is VLAN-based tagging, which requires constant
hardware VLAN filtering. While the VSC73XX family supports provider
bridging, it only supports QinQ without full implementation of 802.1AD.
This means it only allows the doubled 0x8100 TPID.

In the simple port mode, QinQ is enabled to preserve forwarding of
VLAN-tagged frames.

Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-9-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: dsa: vsc73xx: introduce tag 8021q for vsc73xx

This commit introduces a new tagger based on 802.1q tagging.
It's designed for the vsc73xx driver. The VSC73xx family doesn't have
any tag support for the RGMII port, but it could be based on VLANs.

Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Link: https://patch.msgid.link/20240713211620.1125910-8-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: dsa: tag_sja1105: refactor skb->dev assignment to dsa_tag_8021q_find_user()

A new tagging protocol implementation based on tag_8021q is on the
horizon, and it appears that it also has to open-code the complicated
logic of finding a source port based on a VLAN header.

Create a single dsa_tag_8021q_find_user() and make sja1105 call it.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Link: https://patch.msgid.link/20240713211620.1125910-7-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: dsa: tag_sja1105: prefer precise source port info on SJA1110 too

Now that dsa_8021q_rcv() handles better the case where we don't
overwrite the precise source information if it comes from an external
(non-tag_8021q) source, we can now unify the call sequence between
sja1105_rcv() and sja1110_rcv().

This is a preparatory change for creating a higher-level wrapper for the
entire sequence which will live in tag_8021q.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-6-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: dsa: tag_sja1105: absorb entire sja1105_vlan_rcv() into dsa_8021q_rcv()

tag_sja1105 has a wrapper over dsa_8021q_rcv(): sja1105_vlan_rcv(),
which determines whether the packet came from a bridge with
vlan_filtering=1 (the case resolved via
dsa_find_designated_bridge_port_by_vid()), or if it contains a tag_8021q
header.

Looking at a new tagger implementation for vsc73xx, based also on
tag_8021q, it is becoming clear that the logic is needed there as well.
So instead of forcing each tagger to wrap around dsa_8021q_rcv(), let's
merge the logic into the core.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Tested-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-5-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: dsa: tag_sja1105: absorb logic for not overwriting precise info into dsa_8021q_rcv()

In both sja1105_rcv() and sja1110_rcv(), we may have precise source port
information coming from parallel hardware mechanisms, in addition to the
tag_8021q header.

Only sja1105_rcv() has extra logic to not overwrite that precise info
with what's present in the VLAN tag. This is because sja1110_rcv() gets
by, by having a reversed set of checks when assigning skb->dev. When the
source port is imprecise (vbid >=1), source_port and switch_id will be
set to zeroes by dsa_8021q_rcv(), which might be problematic. But by
checking for vbid >= 1 first, sja1110_rcv() fends that off.

We would like to make more code common between sja1105_rcv() and
sja1110_rcv(), and for that, we need to make sure that sja1110_rcv()
also goes through the precise source port preservation logic.

Signed-off-by: Vladimir Oltean <vladimir.oltean@nxp.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Tested-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-4-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: dsa: vsc73xx: Add vlan filtering

This patch implements VLAN filtering for the vsc73xx driver.

After starting VLAN filtering, the switch is reconfigured from QinQ to
a simple VLAN aware mode. This is required because VSC73XX chips do not
support inner VLAN tag filtering.

Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-3-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: dsa: vsc73xx: add port_stp_state_set function

This isn't a fully functional implementation of 802.1D, but
port_stp_state_set is required for a future tag8021q operations.

This implementation handles properly all states, but vsc73xx doesn't
forward STP packets.

Reviewed-by: Linus Walleij <linus.walleij@linaro.org>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Vladimir Oltean <olteanv@gmail.com>
Signed-off-by: Pawel Dembicki <paweldembicki@gmail.com>
Link: https://patch.msgid.link/20240713211620.1125910-2-paweldembicki@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: ti: icssg-prueth: Split out common object into module

icssg_prueth.c and icssg_prueth_sr1.c drivers use multiple common .c
files. These common objects are getting added to multiple modules. As a
result when both drivers are enabled in .config, below warning is seen.

drivers/net/ethernet/ti/Makefile: icssg/icssg_common.o is added to multiple modules: icssg-prueth icssg-prueth-sr1
drivers/net/ethernet/ti/Makefile: icssg/icssg_classifier.o is added to multiple modules: icssg-prueth icssg-prueth-sr1
drivers/net/ethernet/ti/Makefile: icssg/icssg_config.o is added to multiple modules: icssg-prueth icssg-prueth-sr1
drivers/net/ethernet/ti/Makefile: icssg/icssg_mii_cfg.o is added to multiple modules: icssg-prueth icssg-prueth-sr1
drivers/net/ethernet/ti/Makefile: icssg/icssg_stats.o is added to multiple modules: icssg-prueth icssg-prueth-sr1
drivers/net/ethernet/ti/Makefile: icssg/icssg_ethtool.o is added to multiple modules: icssg-prueth icssg-prueth-sr1

Fix this by building a new module (icssg.o) for all the common objects.
Both the driver can then depend on this common module.

Some APIs being exported have emac_ as the prefix which may result into
confusion with other existing APIs with emac_ prefix, to avoid
confusion, rename the APIs being exported with emac_ to icssg_ prefix.

This also fixes below error seen when both drivers are built.
ERROR: modpost: "icssg_queue_pop"
[drivers/net/ethernet/ti/icssg-prueth-sr1.ko] undefined!
ERROR: modpost: "icssg_queue_push"
[drivers/net/ethernet/ti/icssg-prueth-sr1.ko] undefined!

Reported-and-tested-by: Thorsten Leemhuis <linux@leemhuis.info>
Closes: https://lore.kernel.org/oe-kbuild-all/202405182038.ncf1mL7Z-lkp@intel.com/
Fixes: 487f7323f39a ("net: ti: icssg-prueth: Add helper functions to configure FDB")
Reviewed-by: Roger Quadros <rogerq@kernel.org>
Signed-off-by: MD Danish Anwar <danishanwar@ti.com>
Reviewed-by: Sai Krishna <saikrishnag@marvell.com>
Signed-off-by: David S. Miller <davem@davemloft.net>

virtio_net: Fix napi_skb_cache_put warning

After the commit bdacf3e34945 ("net: Use nested-BH locking for
napi_alloc_cache.") was merged, the following warning began to appear:

 WARNING: CPU: 5 PID: 1 at net/core/skbuff.c:1451 napi_skb_cache_put+0x82/0x4b0

  __warn+0x12f/0x340
  napi_skb_cache_put+0x82/0x4b0
  napi_skb_cache_put+0x82/0x4b0
  report_bug+0x165/0x370
  handle_bug+0x3d/0x80
  exc_invalid_op+0x1a/0x50
  asm_exc_invalid_op+0x1a/0x20
  __free_old_xmit+0x1c8/0x510
  napi_skb_cache_put+0x82/0x4b0
  __free_old_xmit+0x1c8/0x510
  __free_old_xmit+0x1c8/0x510
  __pfx___free_old_xmit+0x10/0x10

The issue arises because virtio is assuming it's running in NAPI context
even when it's not, such as in the netpoll case.

To resolve this, modify virtnet_poll_tx() to only set NAPI when budget
is available. Same for virtnet_poll_cleantx(), which always assumed that
it was in a NAPI context.

Fixes: df133f3f9625 ("virtio_net: bulk free tx skbs")
Suggested-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: Breno Leitao <leitao@debian.org>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Reviewed-by: Heng Qi <hengqi@linux.alibaba.com>
Link: https://patch.msgid.link/20240712115325.54175-1-leitao@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'net-phy-bcm5481x-add-support-for-broadr-reach-mode'

Kamil Horák says:

====================
net: phy: bcm5481x: add support for BroadR-Reach mode

PATCH 1 - Add the 10baseT1BRR_Full link mode

PATCH 2 - Add the definitions of LRE registers, necessary to use
   BroadR-Reach modes on the BCM5481x PHY

PATCH 3 - Add brr-mode flag to switch between IEEE802.3 and BroadR-Reach

PATCH 4 - Implementation of the BroadR-Reach modes for the Broadcom
   PHYs
====================

Link: https://patch.msgid.link/20240712150709.3134474-1-kamilh@axis.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: phy: bcm-phy-lib: Implement BroadR-Reach link modes

Implement single-pair BroadR-Reach modes on bcm5481x PHY by Broadcom.
Create set of functions alternative to IEEE 802.3 to handle
configuration of these modes on compatible Broadcom PHYs.
There is only subset of capabilities supported because of limited
collection of hardware available for the development.
For BroadR-Reach capable PHYs, the LRE (Long Reach Ethernet)
alternative register set is handled. Only bcm54811 PHY is verified,
for bcm54810, there is some support possible but untested. There
is no auto-negotiation of the link parameters (called LDS in the
Broadcom terminology, Long-Distance Signaling) for bcm54811.
It should be possible to enable LDS for bcm54810.

Signed-off-by: Kamil Horák (2N) <kamilh@axis.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Link: https://patch.msgid.link/20240712150709.3134474-5-kamilh@axis.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

dt-bindings: ethernet-phy: add optional brr-mode flag

There is a group of PHY chips supporting BroadR-Reach link modes in
a manner allowing for more or less identical register usage as standard
Clause 22 PHY.
These chips support standard Ethernet link modes as well, however, the
circuitry is mutually exclusive and cannot be auto-detected.
The link modes in question are 100Base-T1 as defined in IEEE802.3bw,
based on Broadcom's 1BR-100 link mode, and newly defined 10Base-T1BRR
(1BR-10 in Broadcom documents).

Add optional brr-mode flag to switch the PHY to BroadR-Reach mode.

Signed-off-by: Kamil Horák (2N) <kamilh@axis.com>
Reviewed-by: Rob Herring (Arm) <robh@kernel.org>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Link: https://patch.msgid.link/20240712150709.3134474-4-kamilh@axis.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: phy: bcm54811: Add LRE registers definitions

Add the definitions of LRE registers for Broadcom BCM5481x PHY

Signed-off-by: Kamil Horák (2N) <kamilh@axis.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Link: https://patch.msgid.link/20240712150709.3134474-3-kamilh@axis.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

net: phy: bcm54811: New link mode for BroadR-Reach

Introduce a new link mode necessary for 10 MBit single-pair
connection in BroadR-Reach mode on bcm5481x PHY by Broadcom.
This new link mode, 10baseT1BRR, is known as 1BR10 in the Broadcom
terminology. Another link mode to be used is 1BR100 and it is already
present as 100baseT1, because Broadcom's 1BR100 became 100baseT1
(IEEE 802.3bw).

Signed-off-by: Kamil Horák (2N) <kamilh@axis.com>
Reviewed-by: Florian Fainelli <florian.fainelli@broadcom.com>
Reviewed-by: Andrew Lunn <andrew@lunn.ch>
Link: https://patch.msgid.link/20240712150709.3134474-2-kamilh@axis.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'virtio-net-support-af_xdp-zero-copy'

Xuan Zhuo says:

====================
virtio-net: support AF_XDP zero copy

v5: http://lore.kernel.org/all/20240611114147.31320-1-xuanzhuo@linux.alibaba.com

XDP socket(AF_XDP) is an excellent bypass kernel network framework. The zero
copy feature of xsk (XDP socket) needs to be supported by the driver. The
performance of zero copy is very good. mlx5 and intel ixgbe already support
this feature, This patch set allows virtio-net to support xsk's zerocopy xmit
feature.

At present, we have completed some preparation:

1. vq-reset (virtio spec and kernel code)
2. virtio-core premapped dma
3. virtio-net xdp refactor

So it is time for Virtio-Net to complete the support for the XDP Socket
Zerocopy.

Virtio-net can not increase the queue num at will, so xsk shares the queue with
kernel.

On the other hand, Virtio-Net does not support generate interrupt from driver
manually, so when we wakeup tx xmit, we used some tips. If the CPU run by TX
NAPI last time is other CPUs, use IPI to wake up NAPI on the remote CPU. If it
is also the local CPU, then we wake up napi directly.

This patch set includes some refactor to the virtio-net to let that to support
AF_XDP.

Because there are too many commits, the work of virtio net supporting af-xdp is
split to rx part and tx part. This patch set is for rx part.

So the flag NETDEV_XDP_ACT_XSK_ZEROCOPY is not added, if someone want to test
for af-xdp rx, the flag needs to be adding locally.

ENV: Qemu with vhost-user(polling mode).
Host CPU: Intel(R) Xeon(R) Platinum 8163 CPU @ 2.50GHz

testpmd> show port stats all

 ######################## NIC statistics for port 0 ########################
 RX-packets: 19531092064 RX-missed: 0     RX-bytes: 1093741155584
 RX-errors: 0
 RX-nombuf: 0
 TX-packets: 5959955552 TX-errors: 0     TX-bytes: 371030645664

 Throughput (since last show)
 Rx-pps:   8861574     Rx-bps:  3969985208
 Tx-pps:   8861493     Tx-bps:  3969962736
 ############################################################################

testpmd> show port stats all

  ######################## NIC statistics for port 0  ########################
  RX-packets: 68152727   RX-missed: 0          RX-bytes:  3816552712
  RX-errors: 0
  RX-nombuf:  0
  TX-packets: 68114967   TX-errors: 33216      TX-bytes:  3814438152

  Throughput (since last show)
  Rx-pps:      6333196          Rx-bps:   2837272088
  Tx-pps:      6333227          Tx-bps:   2837285936
  ############################################################################

But AF_XDP consumes more CPU for tx and rx napi(100% and 86%).
====================

Link: https://patch.msgid.link/20240708112537.96291-1-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

virtio_net: xsk: rx: support recv merge mode

Support AF-XDP for merge mode.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-11-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

virtio_net: xsk: rx: support recv small mode

In the process:
1. We may need to copy data to create skb for XDP_PASS.
2. We may need to call xsk_buff_free() to release the buffer.
3. The handle for xdp_buff is difference from the buffer.

If we pushed this logic into existing receive handle(merge and small),
we would have to maintain code scattered inside merge and small (and big).
So I think it is a good choice for us to put the xsk code into an
independent function.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-10-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

virtio_net: xsk: rx: support fill with xsk buffer

Implement the logic of filling rq with XSK buffers.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-9-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

virtio_net: xsk: support wakeup

xsk wakeup is used to trigger the logic for xsk xmit by xsk framework or
user.

Virtio-net does not support to actively generate an interruption, so it
tries to trigger tx NAPI on the local cpu.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-8-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

virtio_net: xsk: bind/unbind xsk for rx

This patch implement the logic of bind/unbind xsk pool to rq.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-7-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

virtio_net: separate receive_mergeable

This commit separates the function receive_mergeable(),
put the logic of appending frag to the skb as an independent function.
The subsequent commit will reuse it.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-6-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

virtio_net: separate receive_buf

This commit separates the function receive_buf(), then we wrap the logic
of handling the skb to an independent function virtnet_receive_done().
The subsequent commit will reuse it.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-5-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

virtio_net: separate virtnet_tx_resize()

This patch separates two sub-functions from virtnet_tx_resize():

* virtnet_tx_pause
* virtnet_tx_resume

Then the subsequent virtnet_tx_reset() can share these two functions.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-4-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

virtio_net: separate virtnet_rx_resize()

This patch separates two sub-functions from virtnet_rx_resize():

* virtnet_rx_pause
* virtnet_rx_resume

Then the subsequent reset rx for xsk can share these two functions.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-3-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

virtio_net: replace VIRTIO_XDP_HEADROOM by XDP_PACKET_HEADROOM

virtio net has VIRTIO_XDP_HEADROOM that is equal to
XDP_PACKET_HEADROOM to calculate the headroom for xdp.

But here we should use the macro XDP_PACKET_HEADROOM from bpf.h to
calculate the headroom for xdp. So here we remove the
VIRTIO_XDP_HEADROOM, and use the XDP_PACKET_HEADROOM to replace it.

Signed-off-by: Xuan Zhuo <xuanzhuo@linux.alibaba.com>
Acked-by: Jason Wang <jasowang@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Link: https://patch.msgid.link/20240708112537.96291-2-xuanzhuo@linux.alibaba.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>

Merge branch 'eliminate-config_nr_cpus-dependency-in-dpaa-eth-and-enable-compile_test-in-fsl_qbman'

Vladimir Oltean says:

====================
Eliminate CONFIG_NR_CPUS dependency in dpaa-eth and enable COMPILE_TEST in fsl_qbman

Breno's previous attempt at enabling COMPILE_TEST for the fsl_qbman
driver (now included here as patch 5/5) triggered compilation warnings
for large CONFIG_NR_CPUS values:
https://lore.kernel.org/all/202406261920.l5pzM1rj-lkp@intel.com/

Patch 1/5 switches two NR_CPUS arrays in the dpaa-eth driver to dynamic
allocation to avoid that warning. There is more NR_CPUS usage in the
fsl-qbman driver, but that looks relatively harmless and I couldn't find
a good reason to change it.

I noticed, while testing, that the driver doesn't actually work properly
with high CONFIG_NR_CPUS values, and patch 2/5 addresses that.

During code analysis, I have identified two places which treat
conditions that can never happen. Patches 3/5 and 4/5 simplify the
probing code - dpaa_fq_setup() - just a little bit.

Finally we have at 5/5 the patch that triggered all of this. There is
an okay from Herbert to take it via netdev, despite it being on soc/qbman:
https://lore.kernel.org/all/Zns%2FeVVBc7pdv0yM@gondor.apana.org.au/

Link to v1:
https://lore.kernel.org/netdev/20240710230025.46487-1-vladimir.oltean@nxp.com/
====================

Link: https://patch.msgid.link/20240713225336.1746343-1-vladimir.oltean@nxp.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>