On 8/3/24 6:15 PM, Anatolij Gustschin wrote:
Conversion to DM_I2C is mandatory, enable DM_I2C
to disable board removal warning. Convert EEPROM
and IT6251 access to use DM_I2C API.
Signed-off-by: Anatolij Gustschin
---
Cc: Marek Vasut
Cc: Fabio Estevam
Cc: Tom Rini
Reviewed-by: Marek Vasut
Hi Sam,
Thanks for working on this, makes switching to upstream dts'es easier
:)
I tested the series on exynos4412-odroid-u2 and exynos5422-odroid-xu4.
One issue for exynos4412-odroid to resolve below, with that fixed
booting from both eMMC and sdcard works on my devices.
On Thu, Jul 25, 2024 at
Hi David,
On Fri, Aug 02, 2024 at 09:19:16PM +0200, David Virag wrote:
> Newer Samsung SoCs (including newer Exynos, ExynosAuto, Google Tensor)
> still use these IPs, or slightly newer versions of it.
>
> Make these drivers available on these platforms by guarding
> EXYNOS4/EXYNOS5 specific code
Hi Jonas,
On Fri, Aug 02, 2024 at 10:42:56PM GMT, Jonas Karlman wrote:
> On 2024-08-02 19:59, Sebastian Reichel wrote:
> > Enable support for the fusb302 USB Type-C controller.
> >
> > This will do early USB PD (power deliver) negotiation, which must happen
> > within 5 seconds after the USB-C co
Hi Jonas,
On Fri, Aug 02, 2024 at 11:04:05PM GMT, Jonas Karlman wrote:
> Hi Sebastian,
>
> On 2024-08-02 19:59, Sebastian Reichel wrote:
> > Since older U-Boot releases do not negotiate USB PD, the kernel
> > DT may not enable the USB-C controller by default to avoid a
> > regression. The plan is
On 02.08.2024 01:55, Chris Morgan wrote:
From: Chris Morgan
The Anbernic RG35XX series of devices are based around an Allwinner
H700 SoC with 1GB of RAM, 2 SD cards, and multiple input buttons.
This bootloader has been tested on the Anbernic RG35XX-2024 and
RG35XX-H, but should be suitable f
On 02.08.2024 01:55, Chris Morgan wrote:
From: Jernej Skrabec
Adjust H616 LPDDR3 DRAM settings to be in line with vendor driver.
Signed-off-by: Jernej Skrabec
Tested-by: Chris Morgan
---
arch/arm/mach-sunxi/dram_sun50i_h616.c | 2 +-
arch/arm/mach-sunxi/dram_timings/h616_lpddr3.c
On 02.08.2024 01:55, Chris Morgan wrote:
From: Jernej Skrabec
Adjust H616 LPDDR4 DRAM settings to be in line with vendor driver.
Signed-off-by: Jernej Skrabec
Tested-by: Chris Morgan
---
arch/arm/mach-sunxi/dram_sun50i_h616.c| 35 +--
.../dram_timings/h616_lpddr4
Hello Mathieu,
after a few months I would like to come back to this patch series from
you.
I recognized you increased from patch v2 to v3 the
CONFIG_SYS_MALLOC_F_LEN from 0x18000 to 0x2.
See v2: https://lists.denx.de/pipermail/u-boot/2024-January/544396.html
diff --git a/configs/imx93-phyboa
From: Maxim Moskalets
Based on rock-3a-rk3568_defconfig.
Tested on v1.31 revision.
Board Specifications:
- Rockchip RK3566
- 1/2/4GB LPDDR4 2112MT/s
- eMMC socket
- uSD card slot
- M.2 2230 Connector
- GbE LAN with POE
- 3.5mm jack with mic
- HDMI 2.0, MIPI DSI/CSI
- USB 3.0 Host, USB 2.0 Host/O
On 01.08.2024 01:58, FUKAUMI Naoki wrote:
> Hi,
>
> please cherry-pick below too
> https://git.kernel.org/pub/scm/linux/kernel/git/devicetree/devicetree-rebasing.git/commit/src/arm64/rockchip/rk3566-rock-3c.dts?h=v6.11-rc1-dts&id=dd40945a1d0e28ae6eaf9da04f8e2dcebf8233ea
>
> On 7/22/24 19:56, Maxim
Conversion to DM_I2C is mandatory, enable DM_I2C
to disable board removal warning.
Signed-off-by: Anatolij Gustschin
---
Cc: Alison Wang
Cc: Fabio Estevam
Cc: Tom Rini
---
configs/vf610twr_defconfig | 2 +-
configs/vf610twr_nand_defconfig | 2 +-
2 files changed, 2 insertions(+), 2 delet
Conversion to DM_I2C is mandatory, enable DM_I2C to fix board
removal warning. Convert EEPROM access to use DM_I2C API.
Signed-off-by: Anatolij Gustschin
---
Cc: Nikita Kiryanov
Cc: Fabio Estevam
Cc: Tom Rini
---
board/compulab/cm_fx6/cm_fx6.c | 3 ++-
board/compulab/common/Makefile | 2 +-
Conversion to DM_I2C is mandatory, enable DM_I2C
to disable board removal warning. Also enable
OF_UPSTREAM since we do not have a default device
tree.
Signed-off-by: Anatolij Gustschin
---
Cc: Paul Kocialkowski
Cc: Tom Rini
---
arch/arm/mach-omap2/omap3/Kconfig | 1 +
configs/sniper_defconfig
Conversion to DM_I2C is mandatory, enable DM_I2C
to disable board removal warning.
Signed-off-by: Anatolij Gustschin
---
Cc: Silvio Fricke
Cc: Fabio Estevam
Cc: Tom Rini
---
board/softing/vining_2000/vining_2000.c | 25 -
configs/vining_2000_defconfig | 4 +-
Conversion to DM_I2C is mandatory, enable DM_I2C
to disable board removal warning. Convert EEPROM
and IT6251 access to use DM_I2C API.
Signed-off-by: Anatolij Gustschin
---
Cc: Marek Vasut
Cc: Fabio Estevam
Cc: Tom Rini
---
board/kosagi/novena/novena.c | 10 +--
board/kosagi/novena/video.c
Conversion to DM_I2C is mandatory, enable DM_I2C
to disable board removal warning.
Signed-off-by: Anatolij Gustschin
---
Cc: Enrico Leto
Cc: Tom Rini
---
configs/rut_defconfig | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configs/rut_defconfig b/configs/rut_defconfig
inde
Conversion to DM_I2C is mandatory, enable DM_I2C
to disable board removal warning.
Signed-off-by: Anatolij Gustschin
---
Cc: Enric Balletbo i Serra
Cc: Tom Rini
---
configs/am335x_sl50_defconfig | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configs/am335x_sl50_defconfig b
Conversion to DM_I2C is mandatory, enable DM_I2C
to disable board removal warning.
Signed-off-by: Anatolij Gustschin
---
Cc: Javier Martínez Canillas
Cc: Tom Rini
---
configs/am335x_igep003x_defconfig | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configs/am335x_igep003x_d
Conversion to DM_I2C is mandatory, enable DM_I2C
to disable board removal warning.
Signed-off-by: Anatolij Gustschin
---
Cc: Nikita Kiryanov
Cc: Tom Rini
---
board/compulab/cm_t43/cm_t43.c | 2 --
configs/cm_t43_defconfig | 3 ++-
2 files changed, 2 insertions(+), 3 deletions(-)
diff --
Conversion to DM_I2C is mandatory, enable DM_I2C
to disable board removal warning.
Signed-off-by: Anatolij Gustschin
---
Cc: Marcin Niestroj
Cc: Tom Rini
---
configs/chiliboard_defconfig | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/configs/chiliboard_defconfig b/configs/
Conversion to DM_I2C is mandatory, rework to remove
use of legacy I2C API.
Signed-off-by: Anatolij Gustschin
---
Cc: Heiko Schocher
Cc: Tom Rini
---
board/bosch/shc/board.c | 22 +++---
configs/am335x_shc_defconfig | 4 +++-
configs/am335x_shc_ict_defconfi
On 31/07/2024 17:58, Vignesh Raghavendra wrote:
> If the function is called with no NAND device attached, then this
> function can return error value, proceeding further ignoring the same
> can cause system crash. This is seen when "mtd list" is run with no NAND
> addon cards connected.
>
> Sig
On Sat, Aug 03, 2024 at 12:41:44PM +, Jonas Karlman wrote:
> The commit b81e31a1e6c5 ("bootstash: Do not provide a default address
> for all") changed a bootstage unstash call to bootstage stash, this
> has resulted in bootstage records stashed in SPL no longer get unstaged
> in U-Boot proper.
The TS433 has 3 buttons, power and copy in the front as well as a reset
pinhole button on the back. The power-button is connected to the embedded
controller while the other two buttons are just gpio connected.
Add the gpio-keys definition for the two buttons we can handle right now.
Tested-by: Uw
The TS433 doesn't provide display output, but the gpu nevertheless can be
used for compute tasks for example.
So there is no reason not to enable it.
Tested-by: Uwe Kleine-König
Signed-off-by: Heiko Stuebner
Link: https://lore.kernel.org/r/20240723195538.1133436-14-he...@sntech.de
[this needs
Add the two supplies for the pmu-io-domains that are defined in the
vendor devicetree for the TS433.
Tested-by: Uwe Kleine-König
Signed-off-by: Heiko Stuebner
Link: https://lore.kernel.org/r/20240723195538.1133436-15-he...@sntech.de
[this needs to be replaced, once the patch hits the devicetree
Fill in the missing pieces for RK809 pmic used on the TS433.
The regulator setup comes from the vendor-devicetree, so without proper
schematics its accuracy is somewhat unclear, but it looks really similar
to all the other rk3568 boards, so follows the reference design it seems.
The one caveat is
The TS433 seems to use a silergy,syr827 regulator for the cpu supply.
At least that is the compatible used in the vendor devicetree, though
it could very well also be another fan53555 clone.
Define the needed regulator node and hook up the cpu-supply to the
cpu cores.
Tested-by: Uwe Kleine-König
The Qnap TS433 is a 4-bay NAS based around the RK3568.
Two SATA bays are connected to the RK3568's own SATA controllers while
the other two are connected to a JMicron SATA controller living on the
PCIe bus.
It provides one 2.5Gb and one 1Gb ethernet port as well as 3 usb ports.
Signed-off-by: He
Enable the tsadc node to allow for temperature measurements of the soc.
Tested-by: Uwe Kleine-König
Signed-off-by: Heiko Stuebner
Link: https://lore.kernel.org/r/20240723195538.1133436-10-he...@sntech.de
[this needs to be replaced, once the patch hits the devicetree-rebasing repo]
Signed-off-b
Add the 4 gpio-controlled LEDs to the Qnap-TS433.
They are meant for individual disk activitivy, but I haven't found a
way for how to connect them to their individual sata slot yet.
Tested-by: Uwe Kleine-König
Signed-off-by: Heiko Stuebner
Link: https://lore.kernel.org/r/20240723195538.1133436-
Add the aliases for the internal network interface as well as the emmc
on the board and make sure the dedicated RTC is always the first one.
The TS433 actually has two rtc devices. One coming from the rk809 pmic
without added functionality and also a dedicated RTC from Mycrocrystal
that is battery
The TS433 uses both pcie controllers for sata and the 2nd network
interface. Set the needed data-lanes in the pcie3 phy and enable
the second pcie controller, as well as remove the bifurcation comment.
Tested-by: Uwe Kleine-König
Signed-off-by: Heiko Stuebner
Link: https://lore.kernel.org/r/2024
Uart0 is connected to an MCU on the board that handles system control
like the fan-speed. So far no driver for it is available though.
Tested-by: Uwe Kleine-König
Signed-off-by: Heiko Stuebner
Link: https://lore.kernel.org/r/20240723195538.1133436-4-he...@sntech.de
[this needs to be replaced, o
As most Rockchip boards do, the TS433 also uses uart2 for its serial
output. Set the correct chosen entry for it.
Tested-by: Uwe Kleine-König
Signed-off-by: Heiko Stuebner
Link: https://lore.kernel.org/r/20240723195538.1133436-6-he...@sntech.de
[this needs to be replaced, once the patch hits th
The TS433 has 4 bays. The last two are accessed via a pci-connected
sata controller, while the first two are accessed via the rk3568's
sata controllers. Enable these two now.
Tested-by: Uwe Kleine-König
Signed-off-by: Heiko Stuebner
Link: https://lore.kernel.org/r/20240723195538.1133436-7-he...@
Enable usb controllers and phys and add regulator infrastructure for the
usb ports on the TS433.
Of course there are no schematics available for the device, so the
regulator information comes from the vendor-devicetree with unknown
accuracy.
Tested-by: Uwe Kleine-König
Signed-off-by: Heiko Stueb
Add the vcc3v3-supply regulator and its link to the pcie controllers.
Tested-by: Uwe Kleine-König
Signed-off-by: Heiko Stuebner
Link: https://lore.kernel.org/r/20240723195538.1133436-2-he...@sntech.de
[this needs to be replaced, once the patch hits the devicetree-rebasing repo]
Signed-off-by:
Currently QNAP builds a series of RK3568- (and RK3588-) based NAS systems.
This series provides support for the 4-bay variant called TS433.
The whole series is pretty similar so adding support for the other
variants will be pretty easy, once device-specific devicetrees land.
The DONOTMERGE label
The timer counter on Rockchip SoCs may be reset in TF-A, this may cause
the bootstage records to be printed out of order and with an incorrect
elapsed time.
Fix this by not sorting the bootstage records.
Before on a Radxa ZERO 3W (RK3566) board:
=> bootstage report
Timer summary in microseco
The commit b81e31a1e6c5 ("bootstash: Do not provide a default address
for all") changed a bootstage unstash call to bootstage stash, this
has resulted in bootstage records stashed in SPL no longer get unstaged
in U-Boot proper. Fix this by changing back to a unstage call.
Fixes: b81e31a1e6c5 ("boo
This series fixes two issues with bootstage observed trying to measure
boot times on my Rockchip aarch64 boards.
First patch fixes a regression in v2024.10-rc1 where bootstage records
from SPL no longer is unstashed.
Second patch change to not sort bootstage records in order to show
correct elaps
Add missing A1/2SYS clock ID just as a reference for OF_UPSTREAM
support. These clocks are not defined and are not usable as current
clock topckgen OPs doesn't support gates.
These special node won't ever be used by uboot hence just add them for
reference.
Signed-off-by: Christian Marangi
---
i
Add missing PERIBUS_SEL clock to match upstream linux clk ID order. Also
convert pericfg to mux + gate implementation as now we have also mux on
top of gates.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7622.c | 22 +-
include/dt-bindings/clock/mt7622-clk.h | 59
Add missing clock PERI_UART4_PD for peri clock gates. This is needed to
match upstream linux clk ID in preparation for OF_UPSTREAM.
Also convert infracfg to mux + gate implementation as now we have mux on
top of gates.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7622.c |
Add missing infra clock MUX1_SEL needed for CPU clock. This is needed to
match the upstream clk ID order in preparation for OF_UPSTREAM.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7622.c | 24 +++-
include/dt-bindings/clock/mt7622-clk.h | 13 +++-
Add missing clock for MAIN_CORE_EN. This is a special clock as it's a
gate for the APMIXED clocks required as a parent for CPU clocks.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7622.c | 29 +++---
include/dt-bindings/clock/mt7622-clk.h | 1 +
2 fil
Move INFRA_TRNG clock to the bottom of the clk ID to match upstream
linux order. This is in preparation of OF_UPSTREAM.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7622.c | 2 +-
include/dt-bindings/clock/mt7622-clk.h | 10 +-
2 files changed, 6 insertions(+), 6
Rename AUDIO_AWB3 to AUDIO_AWB2 to match upstream linux naming in
preparation for OF_UPSTREAM support.
Signed-off-by: Christian Marangi
---
include/dt-bindings/clock/mt7622-clk.h | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/include/dt-bindings/clock/mt7622-clk.h
b/include
Fix broken peri_cgs clock with XTAL parents as they have wrong
definition of the parent type.
Correctly fix them and use CLK_PARENT_XTAL for them.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7622.c | 26 +++---
1 file changed, 15 insertions(+), 11 deletio
These are all the required patches to migrate clk and correctly support
OF_UPSTREAM. This will align the clk index to upstream to support the same
clk implementation with downstream and upstream DTS.
Changes v2:
- Fix typo in commit description
Christian Marangi (8):
clk: mediatek: mt7622: fix
Rename each entry from CK to CLK to match the include in upstream kernel
linux.
Signed-off-by: Christian Marangi
---
arch/arm/dts/mt7986.dtsi | 88 +++---
drivers/clk/mediatek/clk-mt7986.c | 404 -
include/dt-bindings/clock/mt7986-clk.h | 284 -
Convert to infracfg gates + muxes implementation now that it's
supported.
Drop infracfg-ao nodes and rename all infracfg-ao clocks to infracfg.
Signed-off-by: Christian Marangi
---
arch/arm/dts/mt7986.dtsi | 31 ---
drivers/clk/mediatek/clk-mt7986.c | 26 +++
Replace infracfg clk ID with upstream linux version.
The same format is used here with the factor first, then mux and then
gates.
To correctly reference the gates in clk_gate function, define the
gates_offs value in clk_tree now that they are at an offset from mux and
factor.
Signed-off-by: Chri
Move INFRA_TRNG_CK to the bottom of the list to have a 1:1 match with
upstream linux clock ID.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7986.c | 3 +-
include/dt-bindings/clock/mt7986-clk.h | 54 +-
2 files changed, 29 insertions(+), 28 deleti
Comment out CK_TOP_A_TUNER as not used and not defined in upstream
kernel linux. This is to permit support of OF_UPSTREAM and have a 1:1
match with upstream linux clock ID.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7986.c | 3 +-
include/dt-bindings/clock/mt7986-clk.
Now that we can have advanced parent handling for mux, we can drop
spurious topckgen 1/1 factor. This is in preparation to make the clk
ID match the ID in upstream include for mt7986.
Drop the factor entry from mt7986-clk.h and reference to them in
mt7981.dtsi. Muxes and gates are updated to refer
Reorder TOPCKGEN factor ID to put TOP_FACTOR first and then PLL. This is
to match how it's done in upstream kernel linux and in preparation for
OF_UPSTREAM support.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7986.c | 70
include/dt-bindings/clock/mt798
Rename TOPCKGEN factor clock to upstream neaming.
Upstream kernel linux reference the factor clock for apmixedpll with the
"pll" suffix. Align the naming to the upstream naming format in
preparation for OF_UPSTREAM support.
Also rename rtc clock to drop the CB_ as upstream doesn't have that.
Sign
Fix a typo for infra_i2c0_ck where 0 was misspelled as O.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7986.c | 2 +-
include/dt-bindings/clock/mt7986-clk.h | 2 +-
2 files changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/clk/mediatek/clk-mt7986.c
b/drivers
Add missing entry for IPCIE_PIPE_CK infra gate clock. Renumber the clock
order to match the expected offset in the gate array.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7986.c | 3 ++-
include/dt-bindings/clock/mt7986-clk.h | 7 ---
2 files changed, 6 insertions(+)
Now that we can have advanced parent handling for mux, we can drop
spurious infracfg 1/1 factor. This is in preparation to make the clk
ID match the ID in upstream include for mt7986.
Drop the factor entry from mt7986-clk.h and reference to them in
mt7981.dtsi. Muxes and gates are updated to refer
Fix wrong parent for INFRA_ADC_26M_CK as should be INFRA_ADC_FRC_CK
instead of INFRA_CK_F26M. This is to match implementation on upstream
kernel linux.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7986.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drive
Upstream kernel linux clock include use SYSAXI_D2 instead of 66M_MCK.
Rename this clock to the upstream kernel in preparation for support of
OF_UPSTREAM.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7986.c | 20 ++--
include/dt-bindings/clock/mt7986-clk.h
Upstream kernel linux clock include use TOP_XTAL instead of CB_CKSQ_40M.
Rename this clock to the upstream kernel in preparation for support of
OF_UPSTREAM.
Signed-off-by: Christian Marangi
---
arch/arm/dts/mt7986.dtsi | 2 +-
drivers/clk/mediatek/clk-mt7986.c | 72 ++
Fix wrong shift for PCIe clocks. This cause the PCIe port to malfunction
as the gate clocks weren't correctly enabled.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7986.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/drivers/clk/mediatek/clk-mt7986.c
These are all the required patches to migrate clk and correctly support
OF_UPSTREAM. This will align the clk index to upstream to support the same
clk implementation with downstream and upstream DTS.
Christian Marangi (15):
clk: mediatek: mt7986: fix wrong shift for PCIe clocks
clk: mediatek:
Rename each entry from CK to CLK to match the include in upstream kernel
linux.
Signed-off-by: Christian Marangi
---
arch/arm/dts/mt7988.dtsi | 84 +--
drivers/clk/mediatek/clk-mt7988.c | 768 -
include/dt-bindings/clock/mt7988-clk.h | 450 +++-
Convert to infracfg gates + muxes implementation now that it's
supported.
Drop infracfg-ao nodes and rename all infracfg-ao clocks to infracfg.
Signed-off-by: Christian Marangi
---
arch/arm/dts/mt7988.dtsi | 67 ++-
drivers/clk/mediatek/clk-mt7988.c | 24 +--
Replace infracfg clk ID with upstream linux version.
The same format is used here with the factor first, then mux and then
gates.
To correctly reference the gates in clk_gate function, define the
gates_offs value in clk_tree now that they are at an offset from mux and
factor.
Drop any comment th
Comment out infracfg clk not defined in upstream kernel linux clock ID
include. These clock are not used and can be safely commented. Keep them
just to have a reference of their existence.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7988.c | 16 ++--
include/dt-bindings
Now that we can have advanced parent handling for mux, we can drop
spurious topckgen 1/1 factor. This is in preparation to make the clk
ID match the ID in upstream include for mt7988.
Drop the factor entry from mt7988-clk.h and reference to them in
mt7988.dtsi. Muxes and gates are updated to refer
Reorder TOPCKGEN factor ID to put TOP_FACTOR first and then PLL. This is
to match how it's done in upstream kernel linux and in preparation for
OF_UPSTREAM support.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7988.c | 80 +++
include/dt-bindings/clock/mt7988
Rename TOPCKGEN factor clock to upstream neaming.
Upstream kernel linux reference the factor clock for apmixedpll with the
"pll" suffix. Align the naming to the upstream naming format in
preparation for OF_UPSTREAM support.
Also rename rtc clock to drop the CB_ as upstream doesn't have that.
Sign
Now that we can have advanced parent handling for mux, we can drop
spurious infracfg 1/1 factor. This is in preparation to make the clk
ID match the ID in upstream include for mt7988.
Drop the factor entry from mt7988-clk.h and reference to them in
mt7988.dtsi. Muxes and gates are updated to refer
Fix wrong parent for INFRA_PCIE_PERI_26M_CK_P2 as should be
INFRA_PCIE_PERI_26M_CK_P3 instead of INFRA_F26M_O0. This is to match
implementation on upstream kernel linux.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7988.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
Move INFRA_PCIE_PERI_26M_CK_Px clock at top of the infracfg gates
in preparation for support of OF_UPSTREAM to have a 1:1 match with
upstream clock ID.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7988.c | 16 +--
include/dt-bindings/clock/mt7988-clk.h | 168
Upstream kernel linux clock include use TOP_NPU_SEL instead of
TOP_CK_NPU_SEL_CM_TOPS_SEL.
Rename this clock to the upstream kernel in preparation for support of
OF_UPSTREAM.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7988.c | 2 +-
include/dt-bindings/clock/mt7988-clk.
Upstream kernel linux clock include use TOP_DA_SEL instead of
TOP_DA_SELM_XTAL_SEL.
Rename this clock to the upstream kernel in preparation for support of
OF_UPSTREAM.
Signed-off-by: Christian Marangi
---
drivers/clk/mediatek/clk-mt7988.c | 2 +-
include/dt-bindings/clock/mt7988-clk.h | 2 +
Upstream kernel linux clock include use TOP_XTAL instead of CB_CKSQ_40M.
Rename this clock to the upstream kernel in preparation for support of
OF_UPSTREAM.
Signed-off-by: Christian Marangi
---
arch/arm/dts/mt7988.dtsi | 6 +-
drivers/clk/mediatek/clk-mt7988.c | 100
These are all the required patches to migrate clk and correctly support
OF_UPSTREAM. This will align the clk index to upstream to support the same
clk implementation with downstream and upstream DTS.
Christian Marangi (13):
clk: mediatek: mt7988: rename CB_CKSQ_40M to TOP_XTAL
clk: mediatek: m
82 matches
Mail list logo
