| Age | Commit message (Collapse) | Author |
|---|
|
pmk8350 has a random tab character inserted, so remove it.
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Signed-off-by: Nia Espera <nespera@igalia.com>
Link: https://lore.kernel.org/r/20231111-nia-sm8350-for-upstream-v4-3-3a638b02eea5@igalia.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
DMA0 node downstream is specified at 0x900000, so fix the typo. Without
this, enabling any i2c node using DMA0 causes a hang.
Fixes: bc08fbf49bc8 ("arm64: dts: qcom: sm8350: Define GPI DMA engines")
Fixes: 41d6bca799b3 ("arm64: dts: qcom: sm8350: correct DMA controller unit address")
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Signed-off-by: Nia Espera <nespera@igalia.com>
Link: https://lore.kernel.org/r/20231111-nia-sm8350-for-upstream-v4-2-3a638b02eea5@igalia.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
SC8180x APSS Devicetree bindings expect qcom,sc8180x-apss-shared to use
qcom,sdm845-apss-shared fallback:
sc8180x-lenovo-flex-5g.dtb: mailbox@17c00000: compatible: 'oneOf' conditional failed, one must be fixed:
['qcom,sc8180x-apss-shared'] is too short
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Link: https://lore.kernel.org/r/20231111204725.35707-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
The Samsung S6SY761 touchscreen bindings expect AVDD regulator which is
missing in the Sony Xperia PDX225 DTS. There is however pinctrl setting
for in the touchscreen device node with similar name: "avdd", so assume
author wanted to control a fixed regulator with a GPIO.
This fixes dtbs_check warning:
sm6375-sony-xperia-murray-pdx225.dtb: touchscreen@48: 'avdd-supply' is a required property
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20231111164229.63803-6-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Add interrupts to SM6125 DWC3 USB controller, based on downstream/vendor
code of Trinket DTSI from Xiaomi Laurel device, to fix dtbs_check
warnings:
sm6125-xiaomi-laurel-sprout.dtb: usb@4ef8800: 'interrupt-names' is a required property
sm6125-xiaomi-laurel-sprout.dtb: usb@4ef8800: 'oneOf' conditional failed, one must be fixed:
'interrupts' is a required property
'interrupts-extended' is a required property
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Fixes: cff4bbaf2a2d ("arm64: dts: qcom: Add support for SM6125")
Reviewed-by: Marijn Suijten <marijn.suijten@somainline.org>
Link: https://lore.kernel.org/r/20231111164229.63803-5-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Commit 70d1e09ebf19 ("arm64: dts: qcom: sm6115: Use 64 bit addressing")
converted all addresses to 64-bit addressing, but the ARMv7 memory
mapped architected timer bindings expect sizes up to 32-bit. Keep
64-bit addressing but change size of memory mapping to 32-bit
(size-cells=1) and adjust the ranges to match this.
This fixes dtbs_check warnings like:
sm6115p-lenovo-j606f.dtb: timer@f120000: #size-cells:0:0: 1 was expected
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Tested-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Link: https://lore.kernel.org/r/20231111164229.63803-4-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Linux handles both versions, but bindings expect GPIO properties to
have 'gpios' suffix instead of 'gpio':
sa8155p-adp.dtb: pci@1c00000: Unevaluated properties are not allowed ('perst-gpio' was unexpected)
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20231111164229.63803-3-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Linux handles both versions, but bindings expect GPIO properties to
have 'gpios' suffix instead of 'gpio':
sc8180x-primus.dtb: pci@1c10000: Unevaluated properties are not allowed ('perst-gpio', 'wake-gpio' were unexpected)
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Link: https://lore.kernel.org/r/20231111164229.63803-2-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Linux handles both versions, but bindings expect GPIO properties to
have 'gpios' suffix instead of 'gpio':
sc8180x-lenovo-flex-5g.dtb: pci@1c08000: Unevaluated properties are not allowed ('perst-gpio', 'wake-gpio' were unexpected)
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Link: https://lore.kernel.org/r/20231111164229.63803-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Soundwire Devicetree bindings expect the Soundwire controller device
node to be named just "soundwire":
sdm845-db845c.dtb: swm@c85: $nodename:0: 'swm@c85' does not match '^soundwire(@.*)?$'
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Acked-by: Vinod Koul <vkoul@kernel.org>
Link: https://lore.kernel.org/r/20231111095617.16496-2-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
There is no "panic-indicator" default trigger but a property with that
name:
sdm845-db845c.dtb: leds: led-0: Unevaluated properties are not allowed ('linux,default-trigger' was unexpected)
Fixes: 3f72e2d3e682 ("arm64: dts: qcom: Add Dragonboard 845c")
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Link: https://lore.kernel.org/r/20231111095617.16496-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
There is no "panic-indicator" default trigger but a property with that
name:
qrb5165-rb5.dtb: leds: led-user4: Unevaluated properties are not allowed ('linux,default-trigger' was unexpected)
Fixes: b5cbd84e499a ("arm64: dts: qcom: qrb5165-rb5: Add onboard LED support")
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Link: https://lore.kernel.org/r/20231111094623.12476-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
To test out a different GDSC change I wanted to have a USB keypress resume
a system in suspend.
Adding wakeup-source to usb_1 and usb_2 "just works" for me on rb5.
Consistent with qcm2290 and sa8775p add wakeup-source to the dtsi for the
SoC.
Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Link: https://lore.kernel.org/r/20231109004311.2449566-2-bryan.odonoghue@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
The keyboard and touchpad can be used to wake the machine
up from sleep, so mark them as such in the dts file.
Signed-off-by: Steev Klimaszewski <steev@kali.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Link: https://lore.kernel.org/r/20231107-wakeup-source-v2-1-bf1562ef9367@kali.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
The volume down key is controlled by PMIC via the PON hardware on
sa8775p platform, so enable the same for sa8775p-ride.
Signed-off-by: Shazad Hussain <quic_shazhuss@quicinc.com>
Link: https://lore.kernel.org/r/20231107120503.28917-1-quic_shazhuss@quicinc.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
The Qualcomm MDSS bindings expect that DPU opp-table is defined within
DPU node:
sm8350-hdk.dtb: display-subsystem@ae00000: Unevaluated properties are not allowed ('opp-table' was unexpected)
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Link: https://lore.kernel.org/r/20231107103540.27353-3-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Qualcomm MDSS Embedded DisplayPort bindings do not allow
sound-dai-cells:
sc8280xp-lenovo-thinkpad-x13s.dtb: displayport-controller@aea0000: #sound-dai-cells: False schema does not allow [[0]
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Link: https://lore.kernel.org/r/20231107103540.27353-2-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Qualcomm MDSS Embedded DisplayPort bindings do not allow
sound-dai-cells:
sc8180x-primus.dtb: displayport-controller@ae9a000: #sound-dai-cells: False schema does not allow [[0]]
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Link: https://lore.kernel.org/r/20231107103540.27353-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Soundwire Devicetree bindings expect the Soundwire controller device
node to be named just "soundwire":
sm8250-xiaomi-elish-boe.dtb: soundwire-controller@3250000: $nodename:0: 'soundwire-controller@3250000' does not match '^soundwire(@.*)?$'
Reported-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20231107102111.16465-2-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Soundwire Devicetree bindings expect the Soundwire controller device
node to be named just "soundwire":
sc8280xp-lenovo-thinkpad-x13s.dtb: soundwire-controller@3210000: $nodename:0: 'soundwire-controller@3210000' does not match '^soundwire(@.*)?$'
Reported-by: Neil Armstrong <neil.armstrong@linaro.org>
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Neil Armstrong <neil.armstrong@linaro.org>
Link: https://lore.kernel.org/r/20231107102111.16465-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
There is no "multi-ch-bit-off" property in LLCC, according to bindings
and Linux driver:
qdu1000-idp.dtb: system-cache-controller@19200000: 'multi-ch-bit-off' does not match any of the regexes: 'pinctrl-[0-9]+'
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Mukesh Ojha <quic_mojha@quicinc.com>
Link: https://lore.kernel.org/r/20231107080417.16700-2-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
According to bindings and Linux driver there is no
"multi_channel_register" address space for LLCC. The first "reg" entry
is supposed to be llcc0_base since commit 43aa006e074c ("dt-bindings:
arm: msm: Fix register regions used for LLCC banks"):
qdu1000-idp.dtb: system-cache-controller@19200000: reg: [[0, 421527552, 0, 14155776], [0, 438304768, 0, 524288], [0, 572293416, 0, 4]] is too long
qdu1000-idp.dtb: system-cache-controller@19200000: reg-names:0: 'llcc0_base' was expected
qdu1000-idp.dtb: system-cache-controller@19200000: reg-names: ['llcc_base', 'llcc_broadcast_base', 'multi_channel_register'] is too long
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Acked-by: Mukesh Ojha <quic_mojha@quicinc.com>
Link: https://lore.kernel.org/r/20231107080417.16700-1-krzysztof.kozlowski@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Fix the following dt bindings check:
arch/arm64/boot/dts/qcom/sm8450-hdk.dtb: soundwire-controller@31f0000: $nodename:0: 'soundwire-controller@31f0000' does not match '^soundwire(@.*)?$'
from schema $id: http://devicetree.org/schemas/soundwire/qcom,soundwire.yaml#
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20231106-topic-sm8450-upstream-soundwire-bindings-fix-v1-1-41d4844a5a7d@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Fix the following dt bindings check:
arch/arm64/boot/dts/qcom/sm8550-mtp.dtb: soundwire-controller@6ab0000: $nodename:0: 'soundwire-controller@6ab0000' does not match '^soundwire(@.*)?$'
from schema $id: http://devicetree.org/schemas/soundwire/qcom,soundwire.yaml#
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20231106-topic-sm8550-upstream-soundwire-bindings-fix-v1-1-4ded91c805a1@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Add in CAMCC for sc8280xp. The sc8280xp Camera Clock Controller looks
similar to most of the sdmX, smX and now scX controllers.
Signed-off-by: Bryan O'Donoghue <bryan.odonoghue@linaro.org>
Suggested-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Link: https://lore.kernel.org/r/20231026105345.3376-5-bryan.odonoghue@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Enable download mode setting for sm8550 which can help collect
ramdump for this SoC.
Signed-off-by: Mukesh Ojha <quic_mojha@quicinc.com>
Link: https://lore.kernel.org/r/1698253601-11957-4-git-send-email-quic_mojha@quicinc.com
[bjorn: Updated tcsr offset, per Mukesh correction]
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Enable download mode for sm8350 which can help collect
ramdump for this SoC.
Signed-off-by: Mukesh Ojha <quic_mojha@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Link: https://lore.kernel.org/r/1698253601-11957-3-git-send-email-quic_mojha@quicinc.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Enable download mode for sm8250 which can help collect
ramdump for this SoC.
Signed-off-by: Mukesh Ojha <quic_mojha@quicinc.com>
Reviewed-by: Dmitry Baryshkov <dmitry.baryshkov@linaro.org>
Link: https://lore.kernel.org/r/1698253601-11957-2-git-send-email-quic_mojha@quicinc.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Like all other IPQ SoCs, bootloader will collect the system RAM contents
upon crash for the post morterm analysis. If we don't reserve the memory
region used by bootloader, obviously linux will consume it and upon next
boot on crash, bootloader will be loaded in the same region, which will
lead to loose some of the data, sometimes we may miss out critical
information. So lets reserve the region used by the bootloader.
Similarly SBL copies some data into the reserved region and it will be
used in the crash scenario. So reserve 1MB for SBL as well.
While at it, enable the SMEM support along with TCSR mutex.
Signed-off-by: Kathiravan Thirumoorthy <quic_kathirav@quicinc.com>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Link: https://lore.kernel.org/r/20231025-ipq5018-misc-v1-1-7d14fde97fe7@quicinc.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Add the color and function property to the GPIO LED node, which are
missed out in the initial submission.
Signed-off-by: Kathiravan Thirumoorthy <quic_kathirav@quicinc.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20231025-ipq5332-gpio-led-v1-1-0f0f52617648@quicinc.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Add support for wlan-2g LED on GPIO64.
Signed-off-by: Kathiravan Thirumoorthy <quic_kathirav@quicinc.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Link: https://lore.kernel.org/r/20231025-ipq9574-led-v2-1-59b2725697ad@quicinc.com
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
The default for the QCM2290 platform that this board is based on is OTG
mode, however the role detection logic is not hooked up for this board
and the dwc3 driver is configured to not allow role switching from
userspace.
Force this board to host mode as this is the preferred usecase until we
get role switching hooked up.
Fixes: e18771961336 ("arm64: dts: qcom: Add initial QTI RB1 device tree")
Signed-off-by: Caleb Connolly <caleb.connolly@linaro.org>
Reviewed-by: Konrad Dybcio <konrad.dybcio@linaro.org>
Link: https://lore.kernel.org/r/20231025-b4-rb1-usb-host-v1-1-522616c575ef@linaro.org
Signed-off-by: Bjorn Andersson <andersson@kernel.org>
|
|
Tesla FSD is a derivative of Samsung Exynos SoC, thus just like the
others it reuses several devices from older designs. Historically we
kept the old (block's) compatible only. This works fine and there is no
bug here, however guidelines expressed in
Documentation/devicetree/bindings/writing-bindings.rst state that:
1. Compatibles should be specific.
2. We should add new compatibles in case of bugs or features.
Add Tesla FSD compatible specific to be used with an existing fallback.
This will also help reviews of new code using existing DTS as template.
No functional impact on Linux drivers behavior.
Acked-by: Conor Dooley <conor.dooley@microchip.com>
Link: https://lore.kernel.org/r/20231205092229.19135-7-krzysztof.kozlowski@linaro.org
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
|
|
If no cross compiler is specified, try to auto detect one.
Look for various combinations, matching:
powerpc(64(le)?)?(-unknown)?-linux(-gnu)?-
There are more possibilities, but the above is known to find a compiler
on Fedora and Ubuntu (which use linux-gnu-), and also detects the
kernel.org cross compilers (which use linux-).
This allows cross compiling with simply:
# Ubuntu
$ sudo apt install gcc-powerpc-linux-gnu
# Fedora
$ sudo dnf install gcc-powerpc64-linux-gnu
$ make ARCH=powerpc defconfig
$ make ARCH=powerpc -j 4
Inspired by arch/parisc/Makefile.
Acked-by: Segher Boessenkool <segher@kernel.crashing.org>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231206115548.1466874-4-mpe@ellerman.id.au
|
|
If the kernel is being cross compiled, there is no information from
uname on which defconfig is most appropriate, so the Makefile defaults
to ppc64.
However these days almost all distros that support powerpc are little
endian, so it's more likely that defaulting to ppc64le_defconfig will
produce something useful for a user.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231206115548.1466874-3-mpe@ellerman.id.au
|
|
The vdso Makefile adds -U$(ARCH) to CPPFLAGS for the vdso64.lds linker
script. ARCH is always powerpc, so it becomes -Upowerpc, which means
undefine the "powerpc" symbol.
But the 64-bit compiler doesn't define powerpc in the first place,
compare:
$ gcc-5.1.0-nolibc/powerpc64-linux/bin/powerpc64-linux-gcc -m32 -E -dM - </dev/null | grep -w powerpc
#define powerpc 1
$ gcc-5.1.0-nolibc/powerpc64-linux/bin/powerpc64-linux-gcc -m64 -E -dM - </dev/null | grep -w powerpc
$
So there's no need to undefine it for the 64-bit linker script.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231206115548.1466874-2-mpe@ellerman.id.au
|
|
There's no need to use $(ARCH) for references to the arch directory in
the source tree, it is always arch/powerpc.
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231206115548.1466874-1-mpe@ellerman.id.au
|
|
This reverts commit 1abce0580b89 ("powerpc/64s: Fix __pte_needs_flush()
false positive warning")
The previous patch dropped the usage of _PAGE_PRIVILEGED with PAGE_NONE.
Hence this check can be dropped.
Signed-off-by: "Aneesh Kumar K.V (IBM)" <aneesh.kumar@kernel.org>
Reviewed-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231204093638.71503-2-aneesh.kumar@kernel.org
|
|
There used to be a dependency on _PAGE_PRIVILEGED with pte_savedwrite.
But that got dropped by
commit 6a56ccbcf6c6 ("mm/autonuma: use can_change_(pte|pmd)_writable() to replace savedwrite")
With the change in this patch numa fault pte (pte_protnone()) gets mapped as regular user pte
with RWX cleared (no-access) whereas earlier it used to be mapped _PAGE_PRIVILEGED.
Hash fault handling code gets some WARN_ON added in this patch because
those functions are not expected to get called with _PAGE_READ cleared.
commit 18061c17c8ec ("powerpc/mm: Update PROTFAULT handling in the page
fault path") explains the details.
Signed-off-by: "Aneesh Kumar K.V (IBM)" <aneesh.kumar@kernel.org>
Reviewed-by: Christophe Leroy <christophe.leroy@csgroup.eu>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231204093638.71503-1-aneesh.kumar@kernel.org
|
|
In the nestedv2 case, if there is a pending decrementer exception, the
L1 must get the L2's timebase from the L0 to see if the exception should
be cancelled. This adds the overhead of a H_GUEST_GET_STATE call to the
likely case in which the decrementer should not be cancelled.
Avoid this logic for the nestedv2 case.
Signed-off-by: Jordan Niethe <jniethe5@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231201132618.555031-13-vaibhav@linux.ibm.com
|
|
Commit 709236039964 ("KVM: PPC: Reimplement non-SIMD LOAD/STORE
instruction mmio emulation with analyse_instr() input") and
commit 2b33cb585f94 ("KVM: PPC: Reimplement LOAD_FP/STORE_FP instruction
mmio emulation with analyse_instr() input") made
kvmppc_emulate_loadstore() use the results from analyse_instr() for
instruction emulation. In particular the effective address from
analyse_instr() is used for UPDATE type instructions and fact that
op.val is all ready endian corrected is used in the STORE case.
However, these changes now have some negative implications for the
nestedv2 case. For analyse_instr() to determine the correct effective
address, the GPRs must be loaded from the L0. This is not needed as
vcpu->arch.vaddr_accessed is already set. Change back to using
vcpu->arch.vaddr_accessed.
In the STORE case, use kvmppc_get_gpr() value instead of the op.val.
kvmppc_get_gpr() will reload from the L0 if needed in the nestedv2 case.
This means if a byte reversal is needed must now be passed to
kvmppc_handle_store() like in the kvmppc_handle_load() case.
This means the call to kvmhv_nestedv2_reload_ptregs() can be avoided as
there is no concern about op.val being stale. Drop the call to
kvmhv_nestedv2_mark_dirty_ptregs() as without the call to
kvmhv_nestedv2_reload_ptregs(), stale state could be marked as valid.
This is fine as the required marking things dirty is already handled for
the UPDATE case by the call to kvmppc_set_gpr(). For LOADs, it is
handled in kvmppc_complete_mmio_load(). This is called either directly
in __kvmppc_handle_load() if the load can be handled in KVM, or on the
next kvm_arch_vcpu_ioctl_run() if an exit was required.
Signed-off-by: Jordan Niethe <jniethe5@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231201132618.555031-12-vaibhav@linux.ibm.com
|
|
In the nestedv2 case, the L1 may register the L2's VPA with the L0. This
allows the L0 to manage the L2's dispatch count, as well as enable
possible performance optimisations by seeing if certain resources are
not being used by the L2 (such as the PMCs).
Use the H_GUEST_SET_STATE call to inform the L0 of the L2's VPA
address. This can not be done in the H_GUEST_VCPU_RUN input buffer.
Signed-off-by: Jordan Niethe <jniethe5@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231201132618.555031-11-vaibhav@linux.ibm.com
|
|
H_COPY_TOFROM_GUEST is part of the nestedv1 API and so should not be
called by a nestedv2 host. Do not attempt to call it.
Signed-off-by: Jordan Niethe <jniethe5@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231201132618.555031-10-vaibhav@linux.ibm.com
|
|
The msr check in kvmppc_handle_exit_hv() is not needed for nestedv2 hosts,
skip the check to avoid a H_GUEST_GET_STATE hcall.
Signed-off-by: Jordan Niethe <jniethe5@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231201132618.555031-9-vaibhav@linux.ibm.com
|
|
There is no need to inject an external interrupt in
kvmppc_book3s_irqprio_deliver() as the test for BOOK3S_IRQPRIO_EXTERNAL
in kvmhv_run_single_vcpu() before guest entry will raise LPCR_MER if
needed. There is also no need to inject the decrementer interrupt as
this will be raised within the L2 if needed. Avoiding these injections
reduces H_GUEST_GET_STATE hcalls by the L1.
Suggested-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Jordan Niethe <jniethe5@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231201132618.555031-8-vaibhav@linux.ibm.com
|
|
Commit 026728dc5d41 ("KVM: PPC: Book3S HV P9: Inject pending xive
interrupts at guest entry") changed guest entry so that if external
interrupts are enabled, BOOK3S_IRQPRIO_EXTERNAL is not tested for. Test
for this regardless of MSR_EE.
For an L1 host, do not inject an interrupt, but always
use LPCR_MER. If the L0 desires it can inject an interrupt.
Fixes: 026728dc5d41 ("KVM: PPC: Book3S HV P9: Inject pending xive interrupts at guest entry")
Signed-off-by: Nicholas Piggin <npiggin@gmail.com>
[jpn: use kvmpcc_get_msr(), write commit message]
Signed-off-by: Jordan Niethe <jniethe5@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231201132618.555031-7-vaibhav@linux.ibm.com
|
|
LPCR_MER is conditionally set during entry to a guest if there is a
pending external interrupt. In the nestedv2 case, this change is not
being communicated to the L0, which means it is not being set in the L2.
Ensure the updated LPCR value is passed to the L0.
Signed-off-by: Jordan Niethe <jniethe5@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231201132618.555031-6-vaibhav@linux.ibm.com
|
|
kvmhv_copy_tofrom_guest_radix() gets the PID at the start of the
function. If pid is not used, then this is a wasteful H_GUEST_GET_STATE
hcall for nestedv2 hosts. Move the assignment to where pid will be used.
Suggested-by: Nicholas Piggin <npiggin@gmail.com>
Signed-off-by: Jordan Niethe <jniethe5@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231201132618.555031-5-vaibhav@linux.ibm.com
|
|
The check for a hcall coming from userspace is done for KVM-PR. This is
not supported for nestedv2 and the L0 will directly inject the necessary
exception to the L2 if userspace performs a hcall. Avoid checking the
MSR and thus avoid a H_GUEST_GET_STATE hcall in the L1.
Signed-off-by: Jordan Niethe <jniethe5@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231201132618.555031-4-vaibhav@linux.ibm.com
|
|
The kvmppc_get_tb_offset() getter reloads KVMPPC_GSID_TB_OFFSET from the
L0 for nestedv2 host. This is unnecessary as the value does not change.
KVMPPC_GSID_TB_OFFSET also need not be reloaded in
kvmppc_{s,g}et_dec_expires().
Signed-off-by: Jordan Niethe <jniethe5@gmail.com>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
Link: https://msgid.link/20231201132618.555031-3-vaibhav@linux.ibm.com
|
