There are several reasons to remove this from ffs:
  * Memory issues, according to mainline
    32048f4
    ("usb: f_fs: Fix use-after-free for epfile")
  * It is not present on newer (>= 5.4) kernels
  * Smarter ways to debug this are available
  * Possible conflicts with backported mainline commits (see above)

Remove the IPC logging code from ffs and adjust for the commits that relied
on its changes.

This reverts commit d545801 and
e4ed11d.

Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Fixes CVE-2022-0435.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Not sure if a case of commit stealing (see different authors and dates),
but upstream 0d18cda ("usb: dwc3: gadget: Prevent core from
processing stale TRBs") is now replacing 9eaa485 ("usb:
dwc3: gadget: Prevent core from processing stale TRBs").

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Adjust 70051f8 ("usb: gadget: f_rndis functionality driver")
that used spin_lock_irqsave() to use spin_lock() instead, defined in
6d5666f3dc146339167c ("usb: gadget: rndis: add spinlock for rndis response
list").

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Handle IPv6 MTUs the way commit d4d5190 ("xfrm: fix MTU
regression") does, instead of doing downstream's MTU check.

Fixes: 9277207 ("ipv6: remove min MTU check for ipsec tunnels")
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Fixes the following CVEs:
* CVE-2022-23042
* CVE-2022-23041
* CVE-2022-23040
* CVE-2022-23039
* CVE-2022-23038
* CVE-2022-23037
* CVE-2022-23036

Link: https://lore.kernel.org/r/[email protected]
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Conflict resolved when merging upstream d908d27 ("nfc:
st21nfca: Fix potential buffer overflows in EVT_TRANSACTION") and
2c8abaf ("esp: Fix possible buffer overflow in ESP
transformation"), due to ACK and msm changes respectively.

Link: https://lore.kernel.org/r/[email protected]
Tested-by: voidanix <[email protected]>
Signed-off-by: voidanix <[email protected]>

Skipped 9aeb4a5 ("arm64: arch_timer: Add workaround for ARM
erratum 1188873") as it was already present.

Adjusted 0b1c660 ("KVM: arm64: Add templates for BHB
mitigation sequences") due to some extra CPU caps.

Resolved merge conflicts with the following commits:
* 3e39041 ("arm64: Mitigate spectre style branch history side
  channels")
* 6932c82 ("arm64: entry: Add vectors that have the bhb
  mitigation sequences")
* 52d19a0 ("arm64: Add part number for Neoverse N1")
* 786ec17 ("arm64: Add silicon-errata.txt entry for ARM
  erratum 1188873")

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

In commit f045269 ("arm64: Call ARCH_WORKAROUND_2 on
transitions between EL0 and EL1"), the co-author who modified the patch has
left out calls to alternative_cb, stating that code for it was not present.

The code in question is, in fact, present: it nonetheless resulted in a
hardcoded PSCI conduit that used the SMC only.

Pull in the rest of the patch by the original author and use the
alternative_cb macro to take the HVC into account as well.

Fixes: f045269 ("arm64: Call ARCH_WORKAROUND_2 on transitions between EL0 and EL1")
Patch-originally-by: Marc Zyngier <[email protected]>
Signed-off-by: voidanix <[email protected]>

Similarly to commit 7bf76ca ("arm64: Verify PSCI conduit for
ARM_SMCCC_ARCH_WORKAROUND_2"), pull in the rest of the author's patch to
make use of alternative_cb for comparing the ssbd_state.

Fixes: 947ffaa ("arm64: ssbd: Skip apply_ssbd if not using dynamic mitigation")
Patch-originally-by: Marc Zyngier <[email protected]>
Signed-off-by: voidanix <[email protected]>

Conflict with 3fac2cb ("mmc: host: Return an error when
->enable_sdio_irq() ops is missing") resolved.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Conflict with commit 0aa3b63 ("mmc: core: Specify timeouts for
BKOPS and CACHE_FLUSH for eMMC") and 327b668 ("mmc: core:
Default to generic_cmd6_time as timeout in __mmc_switch()") resolved due to
downstream a8b2b7e ("mmc: Merge MMC driver's changes from
kernel msm-4.9 to msm-4.14").

Conflict with commit aaf166f ("Reinstate some of "swiotlb:
rework "fix info leak with DMA_FROM_DEVICE""") also resolved.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Conflict with commit 388bc1e ("dm verity: set
DM_TARGET_IMMUTABLE feature flag") resolved due to downstream
a73c9bc ("BACKPORT: dm verity: add 'check_at_most_once' option
to only validate hashes once").

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Adjusted commit e75b5ea ("mailbox: forward the hrtimer if not
queued and under a lock") for downstream 831c430 ("drivers:
mailbox: fix race resulting in multiple message submission").

Conflict with d221ce5 ("nfc: st21nfca: fix memory leaks in
EVT_TRANSACTION handling") resolved.

Skipped 67d5ad72940875dcc727 ("extcon: Modify extcon device to be created
after driver data is set") due to it rendering our devices unbootable.

Link: https://lore.kernel.org/r/[email protected]
Link: sonyxperiadev/bug_tracker#788
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Tested-by: voidanix <[email protected]>
Signed-off-by: voidanix <[email protected]>

This commit fixes a missing break statement in the switch statement of
the DRM notifier callback function. The missing break caused
fall-through to the next case, which could lead to unintended behavior.

Signed-off-by: Pavel Dubrova <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

With commit 8bc7018 ("random: do not use batches when !crng_ready()"),
calling add_hwgenerator_randomness() from a non-kthread task will trigger
kernel panic. Replace funtion call with add_bootloader_randomness
instead.

Bug: 238701371
Fixes: 8bc7018 ("random: do not use batches when !crng_ready()")
Signed-off-by: Lucas Wei <[email protected]>
Change-Id: I9cedadda4a195266fa015b6479806155cfa96891

In our 4.14.285 import, commit 57a23e7 ("random: pull
add_hwgenerator_randomness() declaration into random.h") moved
add_hwgenerator_randomness() into random.h: adjust the header to make it
compile.

Signed-off-by: voidanix <[email protected]>

This reverts commit b6fde44.

This commit is not only absolute garbage, it is useless as well. Whoever in
CAF-land approved and merged this, please do not hire them.

Signed-off-by: voidanix <[email protected]>

commit e112b03 upstream.

Currently in arm64, FDT is mapped to RO before it's passed to
early_init_dt_scan(). However, there might be some codes
(eg. commit "fdt: add support for rng-seed") that need to modify FDT
during init. Map FDT to RO after early fixups are done.

Signed-off-by: Hsin-Yi Wang <[email protected]>
Reviewed-by: Stephen Boyd <[email protected]>
Reviewed-by: Mike Rapoport <[email protected]>
Signed-off-by: Will Deacon <[email protected]>
[mkbestas: fixed trivial conflicts for 4.14 backport]
Signed-off-by: Michael Bestas <[email protected]>
Signed-off-by: Greg Kroah-Hartman <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Fixes the following CVEs:
* CVE-2022-33744
* CVE-2022-33742
* CVE-2022-33741
* CVE-2022-33740
* CVE-2022-26365

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Conflict with f4a5311 ("xhci: make xhci_handshake timeout for
xhci_reset() adjustable") resolved.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Because of commit f4a5311 ("xhci: make xhci_handshake timeout
for xhci_reset() adjustable"), the reset timeout is now back to a sane
default and also became configurable: in turn it should avoid making the
watchdog timeout again.

All of this made the CAF-issued xhci_handshake_check_state() redundant,
therefore it is best to remove it.

Fixes: 089fc98 ("usb: XHCI: Implement xhci_handshake_check_state() API")
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Fixes CVE-2022-2588

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Conflicts with 967d573 ("usb: dwc3: fix PHY disable sequence")
and 6ce66e3 ("netfilter: nf_conntrack_irc: Fix forged IP
logic") resolved.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Tested-by: voidanix <[email protected]>
Signed-off-by: voidanix <[email protected]>

Conflict with dbd64cf ("netfilter:
nf_conntrack_irc: Tighten matching on DCC message") resolved.

Link: https://lore.kernel.org/r/[email protected]
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Conflict with 6c3da8c ("nilfs2: fix
lockdep warnings in page operations for btree nodes") and
12faed7 ("rpmsg: qcom: glink: replace
strncpy() with strscpy_pad()") resolved.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Conflict with b3275dd ("mmc: core: Fix
kernel panic when remove non-standard SDIO card") and
06035fd ("arm64: errata: Remove AES hwcap
for COMPAT tasks") resolved.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Fixes the following CVEs:
* CVE-2022-3628
* CVE-2022-42895

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Conflict with f60b9ea ("mmc: sdhci: Fix voltage switch delay")
resolved.

Fixes CVE-2022-42896

Link: https://lore.kernel.org/r/[email protected]
Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Fixes the following CVEs:
* CVE-2022-42328
* CVE-2022-42329
* CVE-2022-3643

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Fixes the following CVEs:
* CVE-2023-0266
* CVE-2022-41218

Link: https://lore.kernel.org/r/[email protected]
Signed-off-by: voidanix <[email protected]>

Applied cleanly.

Link: https://lore.kernel.org/r/[email protected]
Tested-by: voidanix <[email protected]>
Signed-off-by: voidanix <[email protected]>

The addresses currently used by our sm8150 DTS are old: so old, in fact,
that kumano ends up being unable to use the modem (no IMEI) and WLAN
when flashing its Android 11 (55.2.A.xxx) provided modem.

Update the addresses to resolve the weird "flash A10 to use A11" mismatch.

Link: sonyxperiadev/bug_tracker#694
Signed-off-by: voidanix <[email protected]>

On Sony's upstream branch, the following 3 commits:
* 9c88d02
* b3e122f
* 24b69cc

Have magically disappeared according to git, possibly due to a force push.

Revert the (now out of tree) changes done to chacha20/random to stay in
sync with upstream and resolve merging conflicts.

Signed-off-by: voidanix <[email protected]>

…y-aosp/LA.UM.7.1.r1

This device has issues with the userspace libdisplayqos as
it will lock up the device due to a totally wrong calculation
of the bus bandwidth and MDP clock rates.

For this reason, sadly say NO to the battery improvements and
ramp up to the sky both busses and MDP clocks while the display
is on.

Signed-off-by: voidanix <[email protected]>