hid_bpf_input_report() is already marked to be used in sleepable context
only. So instead of hammering with timers the device to hopefully get
an available slot where the device is not sending events, we can make
that kfunc wait for the current event to be terminated before it goes in.

This allows to work with the following pseudo code:

in struct_ops/hid_device_event:
  - schedule a bpf_wq, which calls hid_bpf_input_report()
  - once this struct_ops function terminates, hid_bpf_input_report()
    immediately starts before the next event

Link: https://patch.msgid.link/20240626-hid_hw_req_bpf-v2-9-cfd60fb6c79f@kernel.orgAcked-by: Jiri Kosina <jkosina@suse.com>
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>

We add 3 new tests:
- first, we make sure we can prevent the output_report to happen
- second, we make sure that we can detect that a given hidraw client
  was actually doing the request, and for that client only, call ourself
  hid_bpf_hw_output_report(), returning a custom value
- last, we ensure that we can not loop between hooks for
  hid_hw_output_report() and manual calls to hid_bpf_hw_output_report()
  from that same hook

Link: https://patch.msgid.link/20240626-hid_hw_req_bpf-v2-8-cfd60fb6c79f@kernel.orgAcked-by: Jiri Kosina <jkosina@suse.com>
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>

Same story than hid_hw_raw_requests:

This allows to intercept and prevent or change the behavior of
hid_hw_output_report() from a bpf program.

The intent is to solve a couple of use case:
  - firewalling a HID device: a firewall can monitor who opens the hidraw
    nodes and then prevent or allow access to write operations on that
    hidraw node.
  - change the behavior of a device and emulate a new HID feature request

The hook is allowed to be run as sleepable so it can itself call
hid_hw_output_report(), which allows to "convert" one feature request into
another or even call the feature request on a different HID device on the
same physical device.

Link: https://patch.msgid.link/20240626-hid_hw_req_bpf-v2-7-cfd60fb6c79f@kernel.orgAcked-by: Jiri Kosina <jkosina@suse.com>
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>

We add 3 new tests:
- first, we make sure we can prevent the raw_request to happen
- second, we make sure that we can detect that a given hidraw client
  was actually doing the request, and for that client only, call ourself
  hid_bpf_hw_request(), returning a custom value
- last, we ensure that we can not loop between hooks for
  hid_hw_raw_request() and manual calls to hid_bpf_hw_request() from that
  hook

Link: https://patch.msgid.link/20240626-hid_hw_req_bpf-v2-6-cfd60fb6c79f@kernel.orgAcked-by: Jiri Kosina <jkosina@suse.com>
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>

When we attach a sleepable hook to hid_hw_raw_requests, we can (and in
many cases should) call ourself hid_bpf_raw_request(), to actually fetch
data from the device itself.

However, this means that we might enter an infinite loop between
hid_hw_raw_requests hooks and hid_bpf_hw_request() call.

To prevent that, if a hid_bpf_hw_request() call is emitted, we prevent
any new call of this kfunc by storing the information in the context.
This way we can always trace/monitor/filter the incoming bpf requests,
while preventing those loops to happen.

I don't think exposing "from_bpf" is very interesting because while
writing such a bpf program, you need to match at least the report number
and/or the source of the call. So a blind "if there is a
hid_hw_raw_request() call, I'm emitting another one" makes no real
sense.

Link: https://patch.msgid.link/20240626-hid_hw_req_bpf-v2-5-cfd60fb6c79f@kernel.orgAcked-by: Jiri Kosina <jkosina@suse.com>
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>

This allows to intercept and prevent or change the behavior of
hid_hw_raw_request() from a bpf program.

The intent is to solve a couple of use case:
- firewalling a HID device: a firewall can monitor who opens the hidraw
  nodes and then prevent or allow access to write operations on that
  hidraw node.
- change the behavior of a device and emulate a new HID feature request

The hook is allowed to be run as sleepable so it can itself call
hid_bpf_hw_request(), which allows to "convert" one feature request into
another or even call the feature request on a different HID device on the
same physical device.

Link: https://patch.msgid.link/20240626-hid_hw_req_bpf-v2-4-cfd60fb6c79f@kernel.orgAcked-by: Jiri Kosina <jkosina@suse.com>
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>

We want to add sleepable callbacks for hid_hw_raw_request() and
hid_hw_output_report(), but we can not use a plain RCU for those.

Prepare for a SRCU so we can extend HID-BPF.

This changes a little bit how hid_bpf_device_init() behaves, as it may
now fail, so there is a tiny hid-core.c change to accommodate for this.

Link: https://patch.msgid.link/20240626-hid_hw_req_bpf-v2-3-cfd60fb6c79f@kernel.orgAcked-by: Jiri Kosina <jkosina@suse.com>
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>

This allows to know who actually sent what when we process the request
to the device.
This will be useful for a BPF firewall program to allow or not requests
coming from a dedicated hidraw node client.

Link: https://patch.msgid.link/20240626-hid_hw_req_bpf-v2-2-cfd60fb6c79f@kernel.orgAcked-by: Jiri Kosina <jkosina@suse.com>
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>

Looks like if a bpf program gets inserted and then removed,
hdev->bpf.device_data is then allocated, but the loop iterating
over the bpf program is never assigning ret.

This is a problem and also revealed another bug in which only the last
value of ret was checked. This effectively meant than only the last
program in the chain could change the size of the incoming buffer.
Reported-by: Dan Carpenter <dan.carpenter@linaro.org>
Link: https://lore.kernel.org/all/00f7b624-219f-4a05-a7ad-5335f15a41c7@moroto.mountain
Fixes: 4a86220e ("HID: bpf: remove tracing HID-BPF capability")
Link: https://patch.msgid.link/20240626-hid_hw_req_bpf-v2-1-cfd60fb6c79f@kernel.orgAcked-by: Jiri Kosina <jkosina@suse.com>
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>

* spelling fix: XBox -> Xbox, lowercase 'b' as per Microsoft branding
* rephrase: paddle -> paddles, the controller itself has more than one paddle
* rephrase: replace usage of "those" in favor of explicitly making reference
  to the paddles
* grammatical fix: report -> reports, use present tense verb.
* spelling fix: interpret
* consistency: capitalize the first word in bullet points
Signed-off-by: Luis Felipe Hernandez <luis.hernandez093@gmail.com>
Link: https://patch.msgid.link/20240625105553.50830-1-luis.hernandez093@gmail.com
[bentiss: renamed the file into Xbox, not XBox]
Signed-off-by: Benjamin Tissoires <bentiss@kernel.org>

It is useful to change the name, the phys and/or the uniq of a
struct hid_device during .rdesc_fixup().

For example, hid-uclogic.ko changes the uniq to store the firmware version
to differentiate between 2 devices sharing the same PID. In the same
way, changing the device name is useful when the device export 3 nodes,
all with the same name.

Link: https://lore.kernel.org/r/20240608-hid_bpf_struct_ops-v3-16-6ac6ade58329@kernel.orgSigned-off-by: Benjamin Tissoires <bentiss@kernel.org>

The idea is to provide a list of stucts and their editable fields.

Currently no functional changes are introduced here, we will add some
more writeable fields in the next patch.
Acked-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/r/20240608-hid_bpf_struct_ops-v3-15-6ac6ade58329@kernel.orgSigned-off-by: Benjamin Tissoires <bentiss@kernel.org>

Without this helpers like bpf_printk() or bpf_map_update() are not
available, making anything but change of bytes impossible to do.

Link: https://lore.kernel.org/r/20240608-hid_bpf_struct_ops-v3-14-6ac6ade58329@kernel.orgSigned-off-by: Benjamin Tissoires <bentiss@kernel.org>

There is no real reasons to paper over warnings for such small programs.

Link: https://lore.kernel.org/r/20240608-hid_bpf_struct_ops-v3-13-6ac6ade58329@kernel.orgSigned-off-by: Benjamin Tissoires <bentiss@kernel.org>

warning: unused variable ‘tilt’ [-Wunused-variable]
Signed-off-by: Peter Hutterer <peter.hutterer@who-t.net>
Link: https://lore.kernel.org/r/20240608-hid_bpf_struct_ops-v3-12-6ac6ade58329@kernel.orgSigned-off-by: Benjamin Tissoires <bentiss@kernel.org>

This is the current decision we took: we don't provide automatic loading
of HID-BPF by the kernel directly, but rely on an external tool for it.

This tool is currently udev-hid-bpf, so let's make people aware of it.

Link: https://lore.kernel.org/r/20240608-hid_bpf_struct_ops-v3-11-6ac6ade58329@kernel.orgSigned-off-by: Benjamin Tissoires <bentiss@kernel.org>

Now that we are using struct_ops, the docs need to be changed.

Link: https://lore.kernel.org/r/20240608-hid_bpf_struct_ops-v3-10-6ac6ade58329@kernel.orgSigned-off-by: Benjamin Tissoires <bentiss@kernel.org>

I got a weird verifier error with a subprog once, so let's have a test
for it.

Link: https://lore.kernel.org/r/20240608-hid_bpf_struct_ops-v3-9-6ac6ade58329@kernel.orgSigned-off-by: Benjamin Tissoires <bentiss@kernel.org>

We can now rely on struct_ops as we cleared the users in-tree.

Link: https://lore.kernel.org/r/20240608-hid_bpf_struct_ops-v3-8-6ac6ade58329@kernel.orgSigned-off-by: Benjamin Tissoires <bentiss@kernel.org>

Very mechanical:
- Change HID_BPF_DEVICE_EVENT and HID_BPF_RDESC_FIXUP #defines
- add a matching SEC(".struct_ops.link")
- in ArtistPro16Gen2 make the 2 functions static and have a new one
  calling them

Link: https://lore.kernel.org/r/20240608-hid_bpf_struct_ops-v3-7-6ac6ade58329@kernel.orgSigned-off-by: Benjamin Tissoires <bentiss@kernel.org>

We are going to switch over struct_ops, so instead of having to manually
replace all fields one by one, let's have a common place to change it.

Link: https://lore.kernel.org/r/20240608-hid_bpf_struct_ops-v3-6-6ac6ade58329@kernel.orgSigned-off-by: Benjamin Tissoires <bentiss@kernel.org>

This is mostly mechanical: attach_prog is dropped, and
the SEC are converted into struct_ops.

Link: https://lore.kernel.org/r/20240608-hid_bpf_struct_ops-v3-5-6ac6ade58329@kernel.orgSigned-off-by: Benjamin Tissoires <bentiss@kernel.org>

We drop the need for the attach() bpf syscall, but we need to set up
the hid_id field before calling __load().

The .bpf.c part is mechanical: we create one struct_ops per HID-BPF
program, as all the tests are for one program at a time.

Link: https://lore.kernel.org/r/20240608-hid_bpf_struct_ops-v3-4-6ac6ade58329@kernel.orgSigned-off-by: Benjamin Tissoires <bentiss@kernel.org>

We do this implementation in several steps to not have the CI failing:
- first (this patch), we add struct_ops while keeping the existing infra
  available
- then we change the selftests, the examples and the existing in-tree
  HID-BPF programs
- then we remove the existing trace points making old HID-BPF obsolete

There are a few advantages of struct_ops over tracing:
- compatibility with sleepable programs (for hid_hw_raw_request() in
  a later patch)
- a lot simpler in the kernel: it's a simple rcu protected list
- we can add more parameters to the function called without much trouble
- the "attach" is now generic through BPF-core: the caller just needs to
  set hid_id and flags before calling __load().
- all the BPF tough part is not handled in BPF-core through generic
  processing
- hid_bpf_ctx is now only writable where it needs be
Acked-by: Alexei Starovoitov <ast@kernel.org>
Link: https://lore.kernel.org/r/20240608-hid_bpf_struct_ops-v3-3-6ac6ade58329@kernel.orgSigned-off-by: Benjamin Tissoires <bentiss@kernel.org>

no code change, but this way we reduce code duplication and we
can export it later.

Link: https://lore.kernel.org/r/20240608-hid_bpf_struct_ops-v3-2-6ac6ade58329@kernel.orgSigned-off-by: Benjamin Tissoires <bentiss@kernel.org>

Those operations are the ones from HID, not HID-BPF, and I'd like to
reuse hid_bpf_ops as the user facing struct_ops API.

Link: https://lore.kernel.org/r/20240608-hid_bpf_struct_ops-v3-1-6ac6ade58329@kernel.orgSigned-off-by: Benjamin Tissoires <bentiss@kernel.org>

Pull HID fixes from Benjamin Tissoires:

 - fix potential read out of bounds in hid-asus (Andrew Ballance)

 - fix endian-conversion on little endian systems in intel-ish-hid (Arnd
   Bergmann)

 - A couple of new input event codes (Aseda Aboagye)

 - errors handling fixes in hid-nvidia-shield (Chen Ni), hid-nintendo
   (Christophe JAILLET), hid-logitech-dj (José Expósito)

 - current leakage fix while the device is in suspend on a i2c-hid
   laptop (Johan Hovold)

 - other assorted smaller fixes and device ID / quirk entry additions

* tag 'for-linus-2024060801' of git://git.kernel.org/pub/scm/linux/kernel/git/hid/hid:
  HID: Ignore battery for ELAN touchscreens 2F2C and 4116
  HID: i2c-hid: elan: fix reset suspend current leakage
  dt-bindings: HID: i2c-hid: elan: add 'no-reset-on-power-off' property
  dt-bindings: HID: i2c-hid: elan: add Elan eKTH5015M
  dt-bindings: HID: i2c-hid: add dedicated Ilitek ILI2901 schema
  input: Add support for "Do Not Disturb"
  input: Add event code for accessibility key
  hid: asus: asus_report_fixup: fix potential read out of bounds
  HID: logitech-hidpp: add missing MODULE_DESCRIPTION() macro
  HID: intel-ish-hid: fix endian-conversion
  HID: nintendo: Fix an error handling path in nintendo_hid_probe()
  HID: logitech-dj: Fix memory leak in logi_dj_recv_switch_to_dj_mode()
  HID: core: remove unnecessary WARN_ON() in implement()
  HID: nvidia-shield: Add missing check for input_ff_create_memless
  HID: intel-ish-hid: Fix build error for COMPILE_TEST

Merge tag 'kbuild-fixes-v6.10-2' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild

Pull Kbuild fixes from Masahiro Yamada:

 - Fix the initial state of the save button in 'make gconfig'

 - Improve the Kconfig documentation

 - Fix a Kconfig bug regarding property visibility

 - Fix build breakage for systems where 'sed' is not installed in /bin

 - Fix a false warning about missing MODULE_DESCRIPTION()

* tag 'kbuild-fixes-v6.10-2' of git://git.kernel.org/pub/scm/linux/kernel/git/masahiroy/linux-kbuild:
  modpost: do not warn about missing MODULE_DESCRIPTION() for vmlinux.o
  kbuild: explicitly run mksysmap as sed script from link-vmlinux.sh
  kconfig: remove wrong expr_trans_bool()
  kconfig: doc: document behavior of 'select' and 'imply' followed by 'if'
  kconfig: doc: fix a typo in the note about 'imply'
  kconfig: gconf: give a proper initial state to the Save button
  kconfig: remove unneeded code for user-supplied values being out of range

Pull media fixes from Mauro Carvalho Chehab:

 - fixes for the new ipu6 driver (and related fixes to mei csi driver)

 - fix a double debugfs remove logic at mgb4 driver

 - a documentation fix

* tag 'media/v6.10-2' of git://git.kernel.org/pub/scm/linux/kernel/git/mchehab/linux-media:
  media: intel/ipu6: add csi2 port sanity check in notifier bound
  media: intel/ipu6: update the maximum supported csi2 port number to 6
  media: mei: csi: Warn less verbosely of a missing device fwnode
  media: mei: csi: Put the IPU device reference
  media: intel/ipu6: fix the buffer flags caused by wrong parentheses
  media: intel/ipu6: Fix an error handling path in isys_probe()
  media: intel/ipu6: Move isys_remove() close to isys_probe()
  media: intel/ipu6: Fix some redundant resources freeing in ipu6_pci_remove()
  media: Documentation: v4l: Fix ACTIVE route flag
  media: mgb4: Fix double debugfs remove

Pull irq fixes from Ingo Molnar:

 - Fix possible memory leak the riscv-intc irqchip driver load failures

 - Fix boot crash in the sifive-plic irqchip driver caused by recently
   changed boot initialization order

 - Fix race condition in the gic-v3-its irqchip driver

* tag 'irq-urgent-2024-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  irqchip/gic-v3-its: Fix potential race condition in its_vlpi_prop_update()
  irqchip/sifive-plic: Chain to parent IRQ after handlers are ready
  irqchip/riscv-intc: Prevent memory leak when riscv_intc_init_common() fails

Pull x86 fixes from Ingo Molnar:
 "Miscellaneous fixes:

   - Fix kexec() crash if call depth tracking is enabled

   - Fix SMN reads on inaccessible registers on certain AMD systems"

* tag 'x86-urgent-2024-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  x86/amd_nb: Check for invalid SMN reads
  x86/kexec: Fix bug with call depth tracking

Pull perf event fix from Ingo Molnar:
 "Fix race between perf_event_free_task() and perf_event_release_kernel()
  that can result in missed wakeups and hung tasks"

* tag 'perf-urgent-2024-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  perf/core: Fix missing wakeup when waiting for context reference

Pull locking doc fix from Ingo Molnar:
 "Fix typos in the kerneldoc of some of the atomic APIs"

* tag 'locking-urgent-2024-06-08' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip:
  locking/atomic: scripts: fix ${atomic}_sub_and_test() kerneldoc

Merge tag 'mm-hotfixes-stable-2024-06-07-15-24' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm

Pull misc fixes from Andrew Morton:
 "14 hotfixes, 6 of which are cc:stable.

  All except the nilfs2 fix affect MM and all are singletons - see the
  chagelogs for details"

* tag 'mm-hotfixes-stable-2024-06-07-15-24' of git://git.kernel.org/pub/scm/linux/kernel/git/akpm/mm:
  nilfs2: fix nilfs_empty_dir() misjudgment and long loop on I/O errors
  mm: fix xyz_noprof functions calling profiled functions
  codetag: avoid race at alloc_slab_obj_exts
  mm/hugetlb: do not call vma_add_reservation upon ENOMEM
  mm/ksm: fix ksm_zero_pages accounting
  mm/ksm: fix ksm_pages_scanned accounting
  kmsan: do not wipe out origin when doing partial unpoisoning
  vmalloc: check CONFIG_EXECMEM in is_vmalloc_or_module_addr()
  mm: page_alloc: fix highatomic typing in multi-block buddies
  nilfs2: fix potential kernel bug due to lack of writeback flag waiting
  memcg: remove the lockdep assert from __mod_objcg_mlstate()
  mm: arm64: fix the out-of-bounds issue in contpte_clear_young_dirty_ptes
  mm: huge_mm: fix undefined reference to `mthp_stats' for CONFIG_SYSFS=n
  mm: drop the 'anon_' prefix for swap-out mTHP counters

Pull gpio fixes from Bartosz Golaszewski:

 - interrupt handling and Kconfig fixes for gpio-tqmx86

 - add a buffer for storing output values in gpio-tqmx86 as reading back
   the registers always returns the input values

 - add missing MODULE_DESCRIPTION()s to several GPIO drivers

* tag 'gpio-fixes-for-v6.10-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux:
  gpio: add missing MODULE_DESCRIPTION() macros
  gpio: tqmx86: fix broken IRQ_TYPE_EDGE_BOTH interrupt type
  gpio: tqmx86: store IRQ trigger type and unmask status separately
  gpio: tqmx86: introduce shadow register for GPIO output value
  gpio: tqmx86: fix typo in Kconfig label

Pull block fixes from Jens Axboe:

 - Fix for null_blk block size validation (Andreas)

 - NVMe pull request via Keith:
      - Use reserved tags for special fabrics operations (Chunguang)
      - Persistent Reservation status masking fix (Weiwen)

* tag 'block-6.10-20240607' of git://git.kernel.dk/linux:
  null_blk: fix validation of block size
  nvme: fix nvme_pr_* status code parsing
  nvme-fabrics: use reserved tag for reg read/write command

Pull io_uring fixes from Jens Axboe:

 - Fix a locking order issue with setting max async thread workers
   (Hagar)

 - Fix for a NULL pointer dereference for failed async flagged requests
   using ring provided buffers. This doesn't affect the current kernel,
   but it does affect older kernels, and is being queued up for 6.10
   just to make the stable process easier (me)

 - Fix for NAPI timeout calculations for how long to busy poll, and
   subsequently how much to sleep post that if a wait timeout is passed
   in (me)

 - Fix for a regression in this release cycle, where we could end up
   using a partially unitialized match value for io-wq (Su)

* tag 'io_uring-6.10-20240607' of git://git.kernel.dk/linux:
  io_uring: fix possible deadlock in io_register_iowq_max_workers()
  io_uring/io-wq: avoid garbage value of 'match' in io_wq_enqueue()
  io_uring/napi: fix timeout calculation
  io_uring: check for non-NULL file pointer in io_file_can_poll()

Pull btrfs fixes from David Sterba:

 - fix handling of folio private changes.

   The private value holds pointer to our extent buffer structure
   representing a metadata range. Release and create of the range was
   not properly synchronized when updating the private bit which ended
   up in double folio_put, leading to all sorts of breakage

 - fix a crash, reported as duplicate key in metadata, but caused by a
   race of fsync and size extending write. Requires prealloc target
   range + fsync and other conditions (log tree state, timing)

 - fix leak of qgroup extent records after transaction abort

* tag 'for-6.10-rc2-tag' of git://git.kernel.org/pub/scm/linux/kernel/git/kdave/linux:
  btrfs: protect folio::private when attaching extent buffer folios
  btrfs: fix leak of qgroup extent records after transaction abort
  btrfs: fix crash on racing fsync and size-extending write into prealloc

Pull nfsd fix from Chuck Lever:

 - Fix an occasional memory overwrite caused by a fix added in 6.10

* tag 'nfsd-6.10-1' of git://git.kernel.org/pub/scm/linux/kernel/git/cel/linux:
  SUNRPC: Fix loop termination condition in gss_free_in_token_pages()

Pull RISC-V fixes from Palmer Dabbelt:

 - Another fix to avoid allocating pages that overlap with ERR_PTR,
   which manifests on rv32

 - A revert for the badaccess patch I incorrectly picked up an early
   version of

* tag 'riscv-for-linus-6.10-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/riscv/linux:
  Revert "riscv: mm: accelerate pagefault when badaccess"
  riscv: fix overlap of allocated page and PTR_ERR