6.12.17-ti-arm64-r27 bb.org_defconfig
TI SDK: 11.00.06
6.12 TI Delta: https://github.com/RobertCNelson/ti-linux-kernel/compare/1a2d611ee02d63eda3ad6d636e1b0cbf5af87b95...1a22c2fb64485a5dea9a9247eb007ddd4ea8a81d
BBDTBS: BeagleBoard-DeviceTrees@eeb6e572
WIRELESS_REGDB: https://git.kernel.org/pub/scm/linux/kernel/git/wens/wireless-regdb.git/commit/?id=c2f11e2333f0d6dc66141d2c48c522815f6331d6



Signed-off-by: Robert Nelson <robertcnelson@gmail.com>

Signed-off-by: Robert Nelson <robertcnelson@gmail.com>

This reverts commit 3edf588e

.

Signed-off-by: Robert Nelson <robertcnelson@gmail.com>

Signed-off-by: Robert Nelson <robertcnelson@gmail.com>

Signed-off-by: Robert Nelson <robertcnelson@gmail.com>

We want to enable use of the Android binder driver to support
Waydroid, but it should not be built-in as that would waste resources
and increase security attack surface on systems that don't need it.

Export the currently un-exported symbols it depends on.

We want to enable use of the Android ashmem and binder drivers to
support Anbox, but they should not be built-in as that would waste
resources and increase security attack surface on systems that don't
need them.

- Add a MODULE_LICENSE declaration to ashmem
- Change the Makefiles to build each driver as an object with the
  "_linux" suffix (which is what Anbox expects)
- Change config symbol types to tristate

Update:
In upstream commit 721412ed titled "staging: remove ashmem" the ashmem
driver was removed entirely. Secondary commit message:
"The mainline replacement for ashmem is memfd, so remove the legacy
code from drivers/staging/"
Consequently, the ashmem part of this patch has been removed.

Signed-off-by: Ayush Singh <ayush@beagleboard.org>

- provide firmware update support

Signed-off-by: Ayush Singh <ayush@beagleboard.org>

MSPM0L110x microcontrollers (MCUs) are part of the MSP highly-integrated
ultra-low-power 32-bit MSPM0 MCU family based on the enhanced Arm®
Cortex®-M0+ core platform operating at up to 32-MHz frequency.

Signed-off-by: Ayush Singh <ayush@beagleboard.org>

Signed-off-by: Robert Nelson <robertcnelson@gmail.com>

Signed-off-by: Nishanth Menon <nm@ti.com>

When frequency is not exact, the debug information is informative,
but not necessarily informational to the end user. Lower the severity
down.

Fixes: 3e9bee11

 ("ASoC: davinci-mcasp: Restructure the davinci_mcasp_calc_clk_div()")

Signed-off-by: Nishanth Menon <nm@ti.com>

https://openbeagle.org/beagleboard/BeagleBoard-DeviceTrees/-/tree/v6.12.x-Beagle

BeagleBoard-DeviceTrees@eeb6e572


Signed-off-by: Robert Nelson <robertcnelson@gmail.com>

https://git.kernel.org/pub/scm/linux/kernel/git/wens/wireless-regdb.git/commit/?id=c2f11e2333f0d6dc66141d2c48c522815f6331d6


Signed-off-by: Robert Nelson <robertcnelson@gmail.com>

AM62L rev E1 boards dont have TPS65214 but TPS65215. So error out in
probe if we detect TPS65215 on those boards

Going forward E1-1 and future revs would be only board where PMIC would
be supported.

Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>

Enable udma_stop for bcdma_v2 non-cyclic packets. Chan teardown
timeout issue for cyclic packets is still under debug.

Signed-off-by: Sai Sree Kartheek Adivi <s-adivi@ti.com>

Enable udma_reset_rings and refactor k3-udma and k3-udma-am62l drivers.

Signed-off-by: Sai Sree Kartheek Adivi <s-adivi@ti.com>

Add non-tisci implementation for k3_ringacc_ring_reset_sci using direct
register writes.

Fixes: a621a107

 ("PENDING: drivers: soc: ti: k3-ringacc: handle absence of tisci")
Signed-off-by: Sai Sree Kartheek Adivi <s-adivi@ti.com>

am65_cpsw_nuss_init_rx_chns() calls devm_request_irq() first before
registering NAPI via netif_napi_add(). But post devm_request_irq() RX
IRQs are enabled and we may end up calling NAPI poll() callback which
would lead to NULL pointer deref:

 am65-cpsw-nuss 8000000.ethernet: CPTS ver 0x4e8a010d, freq:250000000, add_val:3 pps:0
 am65-cpsw-nuss 8000000.ethernet: set new flow-id-base 96
 Unable to handle kernel NULL pointer dereference at virtual address 0000000000000000
 [...]
 pc : 0x0
 lr : __napi_poll+0x38/0x17c
  0x0
  net_rx_action+0x15c/0x2b8
  handle_softirqs+0x104/0x248
  __do_softirq+0x14/0x20
  ____do_softirq+0x10/0x1c
  call_on_irq_stack+0x24/0x4c
  do_softirq_own_stack+0x1c/0x28
  irq_exit_rcu+0x8c/0xc4
  el1_interrupt+0x38/0x68
  el1h_64_irq_handler+0x18/0x24
  el1h_64_irq+0x64/0x68
  _raw_spin_unlock_irqrestore+0x8/0x44
  request_threaded_irq+0xc8/0x1ac
  devm_request_threaded_irq+0x80/0xfc
  am65_cpsw_nuss_init_rx_chns+0x220/0x45c
  am65_cpsw_nuss_probe+0xb14/0x104c
  platform_probe+0x68/0xc4

Fix this by moving NAPI callback registration to a point before
devm_request_threaded_irq()

Fixes: 82b44cdb

 ("net: ethernet: ti: am65-cpsw: ensure proper channel cleanup in error path")
Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Reviewed-by: Siddharth Vadapalli <s-vadapalli@ti.com>

Enable support for Prolific PL2303 USB Serial single port adapter.

Link: https://lore.kernel.org/all/20250304120320.874617-1-h-salunke@ti.com/


Signed-off-by: Hrushikesh Salunke <h-salunke@ti.com>
Reviewed-by: Siddharth Vadapalli <s-vadapalli@ti.com>

- TPS65214 is a Power Management IC with 3 Buck regulators and 2 LDOs.
- TPS65214 has 2 LDOS and 1 GPO, whereas TPS65219 has 4 LDOs and 2 GPOs.
- TPS65214's LDO1 maps to TPS65219's LDO3.
- A key difference between TPS65215 & TPS65214 are the LDO current and
  voltage output ranges and the configurable options available.
- The remaining features for both devices are the same.

Signed-off-by: Shree Ramamoorthy <s-ramamoorthy@ti.com>

Follow naming convention to have phandle names end with '-pins'.

Link: https://www.kernel.org/doc/Documentation/devicetree/bindings/pinctrl/pinctrl-single.yaml



Signed-off-by: Shree Ramamoorthy <s-ramamoorthy@ti.com>

Enable XDP_SOCKETS.

Signed-off-by: MD Danish Anwar <danishanwar@ti.com>
Signed-off-by: Meghana Malladi <m-malladi@ti.com>

Add native XDP support. We do not support zero copy yet.

Link: https://lore.kernel.org/all/20250305101422.1908370-4-m-malladi@ti.com/


Signed-off-by: Roger Quadros <rogerq@kernel.org>
Signed-off-by: MD Danish Anwar <danishanwar@ti.com>
Signed-off-by: Meghana Malladi <m-malladi@ti.com>

We have different cases for SWDATA (skb, page, cmd, etc)
so it is better to have a dedicated data structure for that.
We can embed the type field inside the struct and use it
to interpret the data in completion handlers.

Link: https://lore.kernel.org/all/20250305101422.1908370-3-m-malladi@ti.com/


Signed-off-by: Roger Quadros <rogerq@kernel.org>
Signed-off-by: MD Danish Anwar <danishanwar@ti.com>
Signed-off-by: Meghana Malladi <m-malladi@ti.com>

This is to prepare for native XDP support.

The page pool API is more faster in allocating pages than
__alloc_skb(). Drawback is that it works at PAGE_SIZE granularity
so we are not efficient in memory usage.
i.e. we are using PAGE_SIZE (4KB) memory for 1.5KB max packet size.

Link: https://lore.kernel.org/all/20250305101422.1908370-2-m-malladi@ti.com/


Signed-off-by: Roger Quadros <rogerq@kernel.org>
Signed-off-by: MD Danish Anwar <danishanwar@ti.com>
Signed-off-by: Meghana Malladi <m-malladi@ti.com>

J721e-SK has handful of different multimedia components such as tidss,
csi, h264 encoder, and h264/h265 decoder that require buffers to be
allocated from CMA region. 512 MB is required in max multistream use case
where all elements are being used.

Signed-off-by: Brandon Brnich <b-brnich@ti.com>

TI's J721e SoC supports Imagination's VXE384 H264 encoder and D5520
H264/H265 decoder.

Add support for both nodes.

Signed-off-by: Brandon Brnich <b-brnich@ti.com>

Video codec driver is expecting CONFIG_VIDEOBUF2_DMA_SG option to be
enabled. Driver is displaying warnings during execution without this option
added. This is just a temporary hack to debug the driver issue further.
Will be removed once a proper solution is found within the driver.

Signed-off-by: Brandon Brnich <b-brnich@ti.com>
Acked-by: Darren Etheridge <detheridge@ti.com>

Introduce debugfs interface to configure cut-thru Port parameters - RX/TX
prio mask.
During probe the debugfs directory is populated for device and every phys
Ports as below
 debug\
 | - <dev name> (8000000.ethernet)
   |- Port1\
	|- <dbgfs files>
   ...
   |- PortN\

For every port following parameters are defined:
 PortN\
 |- [all] disabled [Y/N]
	 Y - port is disabled
 |- [if supported] cut_thru_rx_pri_mask
	 Receive Priority Cut Through Enables
 |- [if supported] cut_thru_tx_pri_mask
	 Transmit Priority Cut Through Enables

cut_thru_rx/tx_pri_mask available only for devices which:
- support "Cut-thru";
- configured in "switch-mode";
and can be configured only when cut-thru is not enabled
  ethtool --show-priv-flags ethX
  cut-thru            : off

Overall configuration sequence example:
 ifconfig ethX down
 devlink dev param set platform/8000000.ethernet name \
  switch_mode value true cmode runtime
 echo 0xc0 > /sys/kernel/debug/8000000.ethernet/PortY/cut_thru_tx_pri_mask
 echo 0xc0 > /sys/kernel/debug/8000000.ethernet/PortY/cut_thru_rx_pri_mask
 ethtool --set-priv-flags ethZ cut-thru on
 ifconfig ethX up

Note. The intended way upstream is to use devlink port params, but this
part of devlink is broken and not easy to fix, so debugfs is used as temp
solution and interface is subject to change.

Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
Signed-off-by: Chintan Vankar <c-vankar@ti.com>

Add basic cut-thru support for AM64 CPSW3g.

The cut-thru through ethtool command:
 ethtool --set-priv-flags eth0 cut-thru on

Once enabled, it will enable cut-thru for RX/TX PRI0 by default.
Note. pri selection not supported yet, PRI0 selected to avoid complex cfg
for now, but will be changed to PRI7.
Port0 cut-thru is not allowed.

It is recommended to use following delays after Link up is detected and
MAC_SL configured - 15us in 10/100 mode and 3us in 1G mode.
Hence, add 40us delay before reading CPSW_PN_SPEED_REG and implement
fall back to manual configuration if link speed auto detection fails.

Signed-off-by: Grygorii Strashko <grygorii.strashko@ti.com>
[vigneshr@ti.com: Replace udelay() with usleep_range()]
Signed-off-by: Vignesh Raghavendra <vigneshr@ti.com>
Signed-off-by: Chintan Vankar <c-vankar@ti.com>

Add PM suspend/resume handling support to k3 r5 remoteproc driver.
The driver registers for pm notifications. On a linux suspend a mailbox
message 'RP_MBOX_SUSPEND_SYSTEM' is sent to the remote core.
a) If the remote core doesn't respond to the SUSPEND message, the
driver sets a device ON constraint and returns.
b) If the remote core responds with 'RP_MBOX_SUSPEND_ACK', the driver
initiates a stop of the remote core and returns from the suspend
handler.
c) If the remote core responds with 'RP_MBOX_SUSPEND_CANCEL', the driver
returns with a -EBUSY error.
d) On the other hand, if the remote core responds with
RP_MBOX_SUSPEND_AUTO the driver simply returns, meaning the remote core
is handling the suspend by itself.

On the Resume path, driver queries DM to determine the state of the
remote core and if it is OFF, turns ON the remote core.

Signed-off-by: Hari Nagalla <hnagalla@ti.com>
Signed-off-by: Markus Schneider-Pargmann <msp@baylibre.com>

Add support to handle LPM suspend/resume. k3 M4 remoteproc driver
registers for pm notifications and binds with k3 common remoteproc
driver's suspend/resume handlers.

Signed-off-by: Hari Nagalla <hnagalla@ti.com>

Add support to handle LPM suspend/resume. k3 DSP remoteproc driver
registers for pm notifications and binds with k3 common remoteproc
driver's suspend/resume handlers.

Signed-off-by: Hari Nagalla <hnagalla@ti.com>

Introduce suspend/resume handling to k3 remoteproc driver. k3 remoteproc
drivers registers for pm notifications. On a linux suspend a mailbox message
'RP_MBOX_SUSPEND_SYSTEM' is sent to the remote core.
a) If the remote core doesn't respond to the SUSPEND message, the
driver sets a device ON constraint and returns.
b) If the remote core responds with 'RP_MBOX_SUSPEND_ACK', the driver
initiates a stop of the remote core and returns from the suspend
handler.
c) If the remote core responds with 'RP_MBOX_SUSPEND_CANCEL', the driver
returns with a -EBUSY error.
c) On the other hand, if the remote core responds with
RP_MBOX_SUSPEND_AUTO the driver simply returns, meaning the remote core
is handling the suspend by itself.

On the Resume path, driver queries DM to determine the state of the
remote core and if it is OFF, turns ON the remote core.

Signed-off-by: Hari Nagalla <hnagalla@ti.com>

Enable PHY mode for OSPI NAND in device tree.

Signed-off-by: Anurag Dutta <a-dutta@ti.com>
Reviewed-by: Udit Kumar <u-kumar1@ti.com>

Enable OSPI NOR PHY for am69-sk in device tree.

Signed-off-by: Prasanth Babu Mantena <p-mantena@ti.com>
Signed-off-by: Anurag Dutta <a-dutta@ti.com>
Reviewed-by: Udit Kumar <u-kumar1@ti.com>