Nodes like pwrkey, resin, iadc, adc-tm, temp-alarm which are the grand
children of spmi_bus node represent the interrupt generating devices but
don't have "interrupt-parent" property.
As per the devicetree spec v0.3, section 2.4:
"The physical wiring of an interrupt source to an interrupt controller is
represented in the devicetree with the interrupt-parent property. Nodes
that represent interrupt-generating devices contain an interrupt-parent
property which has a phandle value that points to the device to which the
device’s interrupts are routed, typically an interrupt controller. If an
interrupt-generating device does not have an interrupt-parent property,
its interrupt parent is assumed to be its devicetree parent."
This clearly says that if the "interrupt-parent" property is absent, then
the immediate devicetree parent will be assumed as the interrupt parent.
But the immediate parents of these nodes are not interrupt controllers
themselves.
This may lead to failure while wiring the interrupt for these nodes by an
operating system. But a few operating systems like Linux, workaround this
issue by walking up the parent nodes until it finds the "interrupt-cells"
property. Then the node that has the "interrupt-cells" property will be
used as the interrupt parent.
But this workaround is not as per the DT spec and is not being implemented
by other operating systems such as OpenBSD.
Hence, fix this issue by adding the "interrupts-extended" property that
explicitly specifies the spmi_bus node as the interrupt parent. Note that
the "interrupts-extended" property is chosen over "interrupt-parent" as it
allows specifying both interrupt parent phandle and interrupt specifiers in
a single property.
Somehow sm8350 got its device nodes not fully sorted. Start reordering
DT nodes by their address. Move apps SMMU, GIC, timer, apps RSC, cpufreq
ADSP and cDSP nodes to the end to the proper position at the end of
/soc/.
arm64: dts: qcom: sc7280: Power herobrine's 3.3 eDP/TS rail more properly
This is the equivalent of commit f5b4811e8758 ("arm64: dts: qcom:
sc7180: Add trogdor eDP/touchscreen regulator off-on-time") and commit 23ff866987de ("arm64: dts: qcom: sc7180: Start the trogdor
eDP/touchscreen regulator on"), but for herobrine instead of trogdor.
The motivations for herobrine are the same as for trogdor.
NOTES:
* Currently for herobrine all boards are eDP, not MIPI. If/when we
have herobrine derivatives that are MIPI they we can evaluate
whether the same off-on-delay makes sense for them. For trogdor we
didn't add the delay to MIPI panels because the problem was found
late and nobody had complained about it. For herobrine defaulting to
assuming the same 500ms makes sense and if we find we need to
optimize later we can.
* Currently there are no oddball herobrine boards like homestar where
the panel really likes to be power cycled. If we have an oddball
board it will need to split the eDP and touchscreen rail anyway
(like homestar did) and we'll have to delete the "regulator-boot-on"
from that board.
arm64: dts: qcom: sc7280: Hook up the touchscreen IO rail on evoker
On older revisions of evoker, the touchscreen was either
non-functional or needed special hardware magic to get it talking
properly. It's been decided that the proper way going forward is to
use L3C to power some buffers on the QCard and then configure the
touchscreens for 1.8V. Let's do that.
Note that this is safe to do even on older revs even if it might not
make the touchscreen work there (because they didn't have a properly
stuffed QCard). As talked about in the patch ("arm64: dts: qcom:
sc7280: On QCard, regulator L3C should be 1.8V") the L3C regulator
didn't go anywhere at all on older revs.
This patch relies on the patch ("HID: i2c-hid: goodix: Add
mainboard-vddio-supply") in order to function properly. Without that
patch this one won't do any harm but it won't actually accomplish its
goal.
arm64: dts: qcom: sc7280: Hook up the touchscreen IO rail on villager
On never revs of sc7280-herobrine-villager (rev2+) the L3C rail is
provided to the touchscreen as the IO voltage rail. Let's add it in
the device tree.
NOTE: Even though this is only really needed on rev2+ villagers (-rev0
had non-functioning touchscreen and -rev1 had some hacky hardware
magic), it doesn't actually hurt to do this for old villager revs. As
talked about in the patch ("arm64: dts: qcom: sc7280: On QCard,
regulator L3C should be 1.8V") the L3C regulator didn't go anywhere at
all on older revs. That means that turning it on for older revs
doesn't hurt other than drawing a tiny bit of extra power. Since -rev0
and -rev1 villagers will never make it to real customers and it's nice
not to have too many old device trees, the better tradeoff seems to be
to enable it everywhere.
arm64: dts: qcom: sc7280: Add 3ms ramp to herobrine's pp3300_left_in_mlb
The "pp3300_left_in_mlb" rail on herobrine eventually connects up to
"vreg_edp_3p3" on the qcard. On several herobrine designs this rail
has been measured to need more than 1ms to turn on.
While technically a herobrine derivative (defined as anyone including
the "herobrine.dtsi") could change the board to make the rail rise
faster or slower, the fact that two boards (evoker and villager) both
measured it as taking more than 1ms implies that it's probably going
to be the norm. Thus, let's add a "regulator-enable-ramp-delay"
straight into the herobrine.dtsi to handle this. If a particular
derivative board needs a faster or slower one then they can override
it, though that feels unlikely.
While we measured something a bit over 1ms, we'll choose 3ms to give
us a tiny bit of margin. This isn't a rail that turns off and on all
the time anyway and 3ms is nothing compared to the total amount of
time to power on a panel.
arm64: dts: qcom: sc7280: On QCard, regulator L3C should be 1.8V
On the first sc7280 QCards the L3C rail was never really used for
anything. Stuffing options on the QCard meant that the QCard itself
didn't use this rail for anything. This rail did get sent to the
mainboard, but no existing mainboards ever did anything with it other
that route it to a testpoint.
On later sc7280 QCards, the L3C rail was repurposed. Instead of being
a (nominally) 3.3V rail, it was decided to make it a 1.8V rail. It is
now provided to the display connector (which might route it to the
touchscreen) and also used to power some buffers relating to
touchscreen IO. This rail is getting the additional tag "ts_avccio",
though some places still refer to it as "vreg_l3c_3p0" despite the
fact that the name now specifies the wrong voltage.
Since it never hurts for this rail to be 1.8V (even on old QCards /
old boards), let's just change it to 1.8V across the board and add the
extra "ts_avccio" moniker as a label in the device tree.
Future patches will start using this rail in their touchscreens.
Jamie Douglass [Thu, 2 Feb 2023 05:48:19 +0000 (16:48 +1100)]
arm64: dts: qcom: msm8992-lg-bullhead: Correct memory overlaps with the SMEM and MPSS memory regions
The memory region reserved by a previous commit (see fixes tag below)
overlaps with the SMEM and MPSS memory regions, causing error messages in
dmesg:
OF: reserved mem: OVERLAP DETECTED!
reserved@5000000 (0x0000000005000000--0x0000000007200000)
overlaps with smem_region@6a00000
(0x0000000006a00000--0x0000000006c00000)
This patch resolves both of these by splitting the previously reserved
memory region into two sections either side of the SMEM region and by
cutting off the second memory region to 0x7000000.
Luca Weiss [Mon, 23 Jan 2023 13:29:51 +0000 (14:29 +0100)]
arm64: dts: qcom: sm6350: Use specific qmpphy compatible
The sc7180 phy compatible works fine for some cases, but it turns out
sm6350 does need proper phy configuration in the driver, so use the
newly added sm6350 compatible.
Because the sm6350 compatible is using the new binding, we need to
change the node quite a bit to match it.
This fixes qmpphy init when no USB cable is plugged in during bootloader
stage.
Luca Weiss [Fri, 20 Jan 2023 13:13:46 +0000 (14:13 +0100)]
arm64: dts: qcom: sm6350: Add CCI nodes
Add nodes for the two CCI blocks found on SM6350.
The first contains two i2c busses and while the second one might also
contains two busses, the downstream kernel only has one configured, and
some boards use the GPIOs for the potential cci1_i2c1 one other
purposes, so leave that one unconfigured.
Andrew Halaney [Mon, 30 Jan 2023 15:48:22 +0000 (09:48 -0600)]
arm64: dts: qcom: sa8540p-ride: Fix some i2c pinctrl settings
Some of the pinctrl groups were invalid for the selected pins. Select
the proper qup group to fix these warnings:
[ 6.523566] sc8280xp-tlmm f100000.pinctrl: invalid group "gpio135" for function "qup15"
[ 6.535042] sc8280xp-tlmm f100000.pinctrl: invalid group "gpio136" for function "qup15"
[ 6.597536] sc8280xp-tlmm f100000.pinctrl: invalid group "gpio158" for function "qup15"
[ 6.597544] sc8280xp-tlmm f100000.pinctrl: invalid group "gpio159" for function "qup15"
[ 6.597991] sc8280xp-tlmm f100000.pinctrl: invalid group "gpio0" for function "qup15"
[ 6.597996] sc8280xp-tlmm f100000.pinctrl: invalid group "gpio1" for function "qup15"
Abel Vesa [Fri, 27 Jan 2023 13:14:41 +0000 (15:14 +0200)]
arm64: dts: qcom: sm8550: Fix the aoss_qmp node name
The proper name for it is power-management. Currently, with the node
name being power-controller, the bindings check fails due to the
property #power-domain-cells missing.
Fixes: ffc50b2d3828 ("arm64: dts: qcom: Add base SM8550 dtsi") Signed-off-by: Abel Vesa <abel.vesa@linaro.org> Reviewed-by: Sai Prakash Ranjan <quic_saipraka@quicinc.com> Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org> Signed-off-by: Bjorn Andersson <andersson@kernel.org> Link: https://lore.kernel.org/r/20230127131441.1157679-1-abel.vesa@linaro.org
Populate the audio devices found on the OnePlus 6 and 6T using the
sdm845-sndcard driver.
Both devices have the earpiece and headphone jack connected to the
WCD9341 codec. The OnePlus 6 uses the MAX98927 speaker codec which is
already supported upstream. The OnePlus 6T uses a currently unsupported
TFA9894 codec.
Two internal microphones are supported, as well as an external headset
mic. Each DAI link is expected to be used for a single device.
Alsa UCM2 configs for this setup can be found here, they are not yet
upstream and include support for call audio which is missing in this
patch
Clock for WCD9340 is coming from the SoC and is the same in all users,
so move it to common file to reduce the code duplication (which still
allows further customizations per board).
Pin configuration fow WCD9340 is the same in all users, so move it to
common file to reduce the code duplication (which still allows further
customizations per board).
arm64: dts: qcom: sdm845: move codec to separate file
Re-organize SDM845 audio codec into separate, audio DTSI which
should be included and customized by the SDM845 boards wanting audio.
The codec node is anyway not a property of the SoC, but the boards.
On all others boards not using audio, keep the Slimbus node disabled as
it is empty.
There is dedicated compatible for Lenovo Yoga C630 sound card
(documented in bindings and used by Linux driver), so use it along with
a generic sound card fallback. The device is actually fully compatible
with the generic one.
Robert Marko [Mon, 23 Jan 2023 10:16:31 +0000 (11:16 +0100)]
arm64: dts: qcom: ipq8074: add QFPROM node
IPQ8074 has efuses like other Qualcomm SoC-s that are required for
determining various HW quirks which will be required later for CPR etc,
so lets add the QFPROM node for start.
Individidual fuses will be added as they are required.
Signed-off-by: Robert Marko <robimarko@gmail.com> Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org> Reviewed-by: Kathiravan T <quic_kathirav@quicinc.com> Signed-off-by: Bjorn Andersson <andersson@kernel.org> Link: https://lore.kernel.org/r/20230123101631.475712-2-robimarko@gmail.com
Johan Hovold [Mon, 23 Jan 2023 10:16:07 +0000 (11:16 +0100)]
arm64: dts: qcom: sm8550: fix USB-DP PHY resets
The USB-DP PHY resets have been switched.
Fixes: 7f7e5c1b037f ("arm64: dts: qcom: sm8550: Add USB PHYs and controller nodes") Signed-off-by: Johan Hovold <johan+linaro@kernel.org> Reviewed-by: Abel Vesa <abel.vesa@linaro.org> Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org> Signed-off-by: Bjorn Andersson <andersson@kernel.org> Link: https://lore.kernel.org/r/20230123101607.2413-1-johan+linaro@kernel.org
Dmitry Baryshkov [Fri, 20 Jan 2023 06:14:15 +0000 (08:14 +0200)]
arm64: dts: qcom: msm8996 switch from RPM_SMD_BB_CLK1 to RPM_SMD_XO_CLK_SRC
The vendor kernel uses RPM_SMD_XO_CLK_SRC clock as an CXO clock rather
than using the RPM_SMD_BB_CLK1 directly. Follow this example and switch
msm8996.dtsi to use RPM_SMD_XO_CLK_SRC clock instead of RPM_SMB_BB_CLK1.
Fixes: 2b8c9c77c268 ("arm64: dts: qcom: msm8996: convert xo_board to RPM_SMD_BB_CLK1") Suggested-by: Konrad Dybcio <konrad.dybcio@linaro.org> Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org> Signed-off-by: Bjorn Andersson <andersson@kernel.org> Link: https://lore.kernel.org/r/20230120061417.2623751-7-dmitry.baryshkov@linaro.org
arm64: dts: qcom: sc7280-idp: drop incorrect properties
The sound card does not expose DAIs and does not use custom qcom
properties, so drop '#sound-dai-cells', 'qcom,msm-mbhc-gnd-swh' and
'qcom,msm-mbhc-hphl-swh':
sc7280-idp.dtb: sound: '#sound-dai-cells', 'qcom,msm-mbhc-gnd-swh', 'qcom,msm-mbhc-hphl-swh' do not match any of the regexes: '^dai-link@[0-9a-f]$', 'pinctrl-[0-9]+'
Reported-by: Srinivasa Rao Mandadapu <quic_srivasam@quicinc.com> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org> Signed-off-by: Bjorn Andersson <andersson@kernel.org> Link: https://lore.kernel.org/r/20230119122205.73372-2-krzysztof.kozlowski@linaro.org
arm64: dts: qcom: sc7280-herobrine-audio-wcd9385: drop incorrect properties
The sound card does not expose DAIs and does not use custom qcom
properties, so drop '#sound-dai-cells', 'qcom,msm-mbhc-gnd-swh' and
'qcom,msm-mbhc-hphl-swh':
sc7280-herobrine-crd.dtb: sound: '#sound-dai-cells', 'qcom,msm-mbhc-gnd-swh', 'qcom,msm-mbhc-hphl-swh' do not match any of the regexes: '^dai-link@[0-9a-f]$', 'pinctrl-[0-9]+'
Reported-by: Srinivasa Rao Mandadapu <quic_srivasam@quicinc.com> Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org> Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org> Signed-off-by: Bjorn Andersson <andersson@kernel.org> Link: https://lore.kernel.org/r/20230119122205.73372-1-krzysztof.kozlowski@linaro.org
Konrad Dybcio [Thu, 19 Jan 2023 10:16:44 +0000 (11:16 +0100)]
arm64: dts: qcom: sm6115: Use 64 bit addressing
SM6115's SMMU uses 36bit VAs, which is a good indicator that we
should increase (dma-)ranges - and by extension #address- and
#size-cells to prevent things from getting lost in translation
(both literally and figuratively). Do so.
Yang Xiwen [Sat, 14 Jan 2023 06:38:46 +0000 (14:38 +0800)]
arm64: dts: qcom: msm8916-thwc: Add initial device trees
This commit adds support for the ufi-001C and uf896 WiFi/LTE dongle made by
Tong Heng Wei Chuang based on MSM8916.
uf896 is another variant for the usb stick. The board design
differs by using different gpios for the keys and leds.
Note: The original firmware does not support 64-bit OS. It is necessary
to flash 64-bit TZ firmware to boot arm64.
Currently supported:
- All CPU cores
- Buttons
- LEDs
- Modem
- SDHC
- USB Device Mode
- UART
arm64: dts: qcom: sm8550: add GPR and LPASS pin controller
Add the ADSP GPR (Generic Packet Router) and LPASS LPI (Low Power Audio
SubSystem Low Power Island) pin controller nodes used as part of audio
subsystem on SM8550.
Rajendra Nayak [Fri, 16 Dec 2022 11:29:18 +0000 (16:59 +0530)]
arm64: dts: qcom: sc7280: Add a herobrine CRD Pro SKU
Some of the qualcomm qcard based herobrine devices can come with
a Pro variant of the chipset on the qcard. Such Pro qcards have
the smps9 from pm8350c ganged up with smps7 and smps8, so add a
.dtsi for pro skus that deletes the smps9 node and include it from
the new dts for the CRD Pro
Melody Olvera [Thu, 12 Jan 2023 21:07:21 +0000 (13:07 -0800)]
arm64: dts: qcom: Add base QDU1000/QRU1000 DTSIs
Add the base DTSI files for QDU1000 and QRU1000 SoCs, including base
descriptions of CPUs, GCC, RPMHCC, QUP, TLMM, and interrupt-controller
to boot to shell with console on these SoCs.
Konrad Dybcio [Fri, 20 Jan 2023 21:00:59 +0000 (22:00 +0100)]
arm64: dts: qcom: sm8350: Add mdss_ prefix to DSIn out labels
Add the mdss_ prefix to DSIn labels, so that the hardware blocks can
be organized near each other while retaining the alphabetical order
in device DTs when referencing by label.
Current PCIe QMP PHY output name were changed in ("arm64: dts: qcom: Fix
IPQ8074 PCIe PHY nodes") however it did not account for the fact that GCC
driver is relying on the old names to match them as they are being used as
the parent for the gcc_pcie0_pipe_clk and gcc_pcie1_pipe_clk.
This broke parenting as GCC could not find the parent clock, so fix it by
changing to the names that driver is expecting.
Robert Marko [Fri, 13 Jan 2023 16:44:48 +0000 (17:44 +0100)]
arm64: dts: qcom: ipq8074: fix Gen3 PCIe node
IPQ8074 comes in 2 silicon versions:
* v1 with 2x Gen2 PCIe ports and QMP PHY-s
* v2 with 1x Gen3 and 1x Gen2 PCIe ports and QMP PHY-s
v2 is the final and production version that is actually supported by the
kernel, however it looks like PCIe related nodes were added for the v1 SoC.
Finish the PCIe fixup by using the correct compatible, adding missing ATU
register space, declaring max-link-speed, use correct ranges, add missing
clocks and resets.
projects / thirdparty / linux.git / log
