phy: marvell: a3700-comphy: Fix hardcoded array size

Replace hardcoded 'gbe_phy_init' array size by explicit one.

Fixes: 934337080c6c ("phy: marvell: phy-mvebu-a3700-comphy: Add native kernel implementation")
Signed-off-by: Mikhail Kobuk <m.kobuk@ispras.ru>
Link: https://lore.kernel.org/r/20240321164734.49273-2-m.kobuk@ispras.ru
Signed-off-by: Vinod Koul <vkoul@kernel.org>

phy: marvell: a3700-comphy: Fix out of bounds read

There is an out of bounds read access of 'gbe_phy_init_fix[fix_idx].addr'
every iteration after 'fix_idx' reaches 'ARRAY_SIZE(gbe_phy_init_fix)'.

Make sure 'gbe_phy_init[addr]' is used when all elements of
'gbe_phy_init_fix' array are handled.

Found by Linux Verification Center (linuxtesting.org) with SVACE.

Fixes: 934337080c6c ("phy: marvell: phy-mvebu-a3700-comphy: Add native kernel implementation")
Signed-off-by: Mikhail Kobuk <m.kobuk@ispras.ru>
Reviewed-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/r/20240321164734.49273-1-m.kobuk@ispras.ru
Signed-off-by: Vinod Koul <vkoul@kernel.org>

phy: constify of_phandle_args in xlate

The xlate callbacks are supposed to translate of_phandle_args to proper
provider without modifying the of_phandle_args. Make the argument
pointer to const for code safety and readability.

Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Thierry Reding <treding@nvidia.com>
Acked-by: Linus Walleij <linus.walleij@linaro.org>
Acked-by: Florian Fainelli <florian.fainelli@broadcom.com> #Broadcom
Link: https://lore.kernel.org/r/20240217093937.58234-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

phy: Explicitly include correct DT includes

The DT of_device.h and of_platform.h date back to the separate
of_platform_bus_type before it as merged into the regular platform bus.
As part of that merge prepping Arm DT support 13 years ago, they
"temporarily" include each other. They also include platform_device.h
and of.h. As a result, there's a pretty much random mix of those include
files used throughout the tree. In order to detangle these headers and
replace the implicit includes with struct declarations, users need to
explicitly include the correct includes.

Signed-off-by: Rob Herring <robh@kernel.org>
Acked-by: Marc Kleine-Budde <mkl@pengutronix.de> # for drivers/phy/phy-can-transceiver.c
Acked-by: Heiko Stuebner <heiko@sntech.de>
Acked-by: Sergio Paracuellos <sergio.paracuellos@gmail.com>
Link: https://lore.kernel.org/r/20230714174841.4061919-1-robh@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

phy: marvell: phy-mvebu-a3700-comphy: Reset COMPHY registers before USB 3.0 power on

Turris MOX board with older ARM Trusted Firmware version v1.5 is not able
to detect any USB 3.0 device connected to USB-A port on Mox-A module after
commit 0a6fc70d76bd ("phy: marvell: phy-mvebu-a3700-comphy: Remove broken
reset support"). On the other hand USB 2.0 devices connected to the same
USB-A port are working fine.

It looks as if the older firmware configures COMPHY registers for USB 3.0
somehow incompatibly for kernel driver. Experiments show that resetting
COMPHY registers via setting SFT_RST auto-clearing bit in COMPHY_SFT_RESET
register fixes this issue.

Reset the COMPHY in mvebu_a3700_comphy_usb3_power_on() function as a first
step after selecting COMPHY lane and USB 3.0 function. With this change
Turris MOX board can successfully detect USB 3.0 devices again.

Before the above mentioned commit this reset was implemented in PHY reset
method, so this is the reason why there was no issue with older firmware
version then.

Fixes: 0a6fc70d76bd ("phy: marvell: phy-mvebu-a3700-comphy: Remove broken reset support")
Reported-by: Marek Behún <kabel@kernel.org>
Signed-off-by: Pali Rohár <pali@kernel.org>
Tested-by: Shin'ichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Link: https://lore.kernel.org/r/20220920121154.30115-1-pali@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

phy: marvell: phy-mvebu-a3700-comphy: Remove broken reset support

Reset support for SATA PHY is somehow broken and after calling it, kernel
is not able to detect and initialize SATA disk Samsung SSD 850 EMT0 [1].

Reset support was introduced in commit 934337080c6c ("phy: marvell:
phy-mvebu-a3700-comphy: Add native kernel implementation") as part of
complete rewrite of this driver. v1 patch series of that commit [2] did
not contain reset support and was tested that is working fine with
Ethernet, SATA and USB PHYs without issues too.

So for now remove broken reset support and change implementation of
power_off callback to power off all functions on specified lane (and not
only selected function) because during startup kernel does not know which
function was selected and configured by bootloader. Same logic was used
also in v1 patch series of that commit.

This change fixes issues with initialization of SATA disk Samsung SSD 850
and disk is working again, like before mentioned commit.

Once problem with PHY reset callback is solved its functionality could be
re-introduced. But for now it is unknown why it does not work.

[1] - https://lore.kernel.org/r/20220531124159.3e4lgn2v462irbtz@shindev/
[2] - https://lore.kernel.org/r/20211028184242.22105-1-kabel@kernel.org/

Reported-by: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Fixes: 934337080c6c ("phy: marvell: phy-mvebu-a3700-comphy: Add native kernel implementation")
Cc: stable@vger.kernel.org # v5.18+
Signed-off-by: Pali Rohár <pali@kernel.org>
Tested-by: Shinichiro Kawasaki <shinichiro.kawasaki@wdc.com>
Link: https://lore.kernel.org/r/20220829083046.15082-1-pali@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

phy: marvell: phy-mvebu-a3700-comphy: Add native kernel implementation

Remove old RPC implementation and add a new native kernel implementation.

The old implementation uses ARM SMC API to issue RPC calls to ARM Trusted
Firmware which provides real implementation of PHY configuration.

But older versions of ARM Trusted Firmware do not provide this PHY
configuration functionality, simply returning: operation not supported; or
worse, some versions provide the configuration functionality incorrectly.

For example the firmware shipped in ESPRESSObin board has this older
version of ARM Trusted Firmware and therefore SATA, USB 3.0 and PCIe
functionality do not work with newer versions of Linux kernel.

Due to the above reasons, the following commits were introduced into Linux,
to workaround these issues by ignoring -EOPNOTSUPP error code from
phy-mvebu-a3700-comphy driver function phy_power_on():

commit 45aefe3d2251 ("ata: ahci: mvebu: Make SATA PHY optional for Armada
3720")
commit 3241929b67d2 ("usb: host: xhci: mvebu: make USB 3.0 PHY optional for
Armada 3720")
commit b0c6ae0f8948 ("PCI: aardvark: Fix initialization with old Marvell's
Arm Trusted Firmware")

Replace this RPC implementation with proper native kernel implementation,
which is independent on the firmware. Never return -EOPNOTSUPP for proper
arguments.

This should solve multiple issues with real-world boards, where it is not
possible or really inconvenient to change the firmware. Let's eliminate
these issues.

This implementation is ported directly from Armada 3720 comphy driver found
in newest version of ARM Trusted Firmware source code, but with various
fixes of register names, some added comments, some refactoring due to the
original code not conforming to kernel standards. Also PCIe mode poweroff
support was added here, and PHY reset support. These changes are also going
to be sent to ARM Trusted Firmware.

[ Pali did the porting from ATF.
I (Marek) then fixed some register names, some various other things,
added some comments and refactored the code to kernel standards. Also
fixed PHY poweroff and added PHY reset. ]

Signed-off-by: Pali Rohár <pali@kernel.org>
Acked-by: Miquel Raynal <miquel.raynal@bootlin.com>
Signed-off-by: Marek Behún <kabel@kernel.org>
Link: https://lore.kernel.org/r/20220203214444.1508-3-kabel@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

phy: marvell: phy-mvebu-a3700-comphy: Remove port from driver configuration

Port number is encoded into argument for SMC call. It is zero for SATA,
PCIe and also both USB 3.0 PHYs. It is non-zero only for Ethernet PHY
(incorrectly called SGMII) on lane 0. Ethernet PHY on lane 1 also uses zero
port number.

So construct "port" bits for SMC call argument can be constructed directly
from PHY type and lane number.

Change driver code to always pass zero port number for non-ethernet PHYs
and for ethernet PHYs determinate port number from lane number. This
simplifies the driver.

As port number from DT PHY configuration is not used anymore, remove whole
driver code which parses it. This also simplifies the driver.

Signed-off-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Marek Behún <kabel@kernel.org>
Reviewed-by: Miquel Raynal <miquel.raynal@bootlin.com>
Link: https://lore.kernel.org/r/20220203214444.1508-2-kabel@kernel.org
Signed-off-by: Vinod Koul <vkoul@kernel.org>

phy: marvell: phy-mvebu-a3700-comphy: Remove unsupported modes

Armada 3700 does not support RXAUI, XFI and neither SFI. Remove unused
macros for these unsupported modes.

Signed-off-by: Pali Rohár <pali@kernel.org>
Fixes: 9695375a3f4a ("phy: add A3700 COMPHY support")
Signed-off-by: David S. Miller <davem@davemloft.net>

phy: marvell: phy-mvebu-a3700-comphy: Rename HS-SGMMI to 2500Base-X

Comphy phy mode 0x3 is incorrectly named. It is not SGMII but rather
2500Base-X mode which runs at 3.125 Gbps speed.

Rename macro names and comments to 2500Base-X.

Signed-off-by: Pali Rohár <pali@kernel.org>
Fixes: 9695375a3f4a ("phy: add A3700 COMPHY support")
Signed-off-by: David S. Miller <davem@davemloft.net>

phy: marvell: comphy: Convert internal SMCC firmware return codes to errno

Driver ->power_on and ->power_off callbacks leaks internal SMCC firmware
return codes to phy caller. This patch converts SMCC error codes to
standard linux errno codes. Include file linux/arm-smccc.h already provides
defines for SMCC error codes, so use them instead of custom driver defines.
Note that return value is signed 32bit, but stored in unsigned long type
with zero padding.

Tested-by: Tomasz Maciej Nowak <tmn505@gmail.com>
Link: https://lore.kernel.org/r/20200902144344.16684-2-pali@kernel.org
Signed-off-by: Pali Rohár <pali@kernel.org>
Signed-off-by: Lorenzo Pieralisi <lorenzo.pieralisi@arm.com>
Reviewed-by: Rob Herring <robh@kernel.org>

phy: mvebu-a3700-comphy: Inform users if their firmware is too old

PHY configuration has been implemented in the firmware and accessed
through SMC calls. In the past, it worked magically if the bootloader
was correctly doing the initializations.

With up-to-date bindings, the kernel will need a recent firmware in
order to do the initializations himself (we assume people must update
their firmware along with their kernel).

People might not understand why IPs that were working correctly before
stopped to be probed suddendly. In this case, let's advise the users
to update their firmware with a visual warning.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Signed-off-by: Kishon Vijay Abraham I <kishon@ti.com>

phy: marvell: phy-mvebu-a3700-comphy: Add of_node_put() before return

Each iteration of for_each_available_child_of_node puts the previous
node, but in the case of a return from the middle of the loop, there is
no put, thus causing a memory leak. Hence add an of_node_put before the
return in two places.
Issue found with Coccinelle.

Signed-off-by: Nishka Dasgupta <nishkadg.linux@gmail.com>
Signed-off-by: Kishon Vijay Abraham I <kishon@ti.com>

phy: add A3700 COMPHY support

Add a driver to support COMPHY, a hardware block providing shared
serdes PHYs on Marvell Armada 3700. This driver uses SMC calls and
rely on having an up-to-date firmware.

SATA, PCie and USB3 host mode have been tested successfully with an
ESPRESSObin. (HS)SGMII mode cannot be tested with this platform.

Evan worked on the original driver structure and Grzegorz on the SMC
calls rework. The structure of this driver has been copied from
Antoine Tenart work on CP110 COMPHY driver.

Signed-off-by: Miquel Raynal <miquel.raynal@bootlin.com>
Co-developed-by: Evan Wang <xswang@marvell.com>
Signed-off-by: Evan Wang <xswang@marvell.com>
Co-developed-by: Grzegorz Jaszczyk <jaz@semihalf.com>
Signed-off-by: Grzegorz Jaszczyk <jaz@semihalf.com>
Signed-off-by: Kishon Vijay Abraham I <kishon@ti.com>