- 19 May, 2022 4 commits
-
-
Greg Kroah-Hartman authored
Merge tag 'soundwire-5.19-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/soundwire into char-misc-next Vinod writes: soundwire updates for 5.19-rc1 - Support for v1.6.0 Qualcomm controllers - Bunch of pm updates by Intel for peripheral attachment and system pm etc * tag 'soundwire-5.19-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vkoul/soundwire: soundwire: qcom: adjust autoenumeration timeout soundwire: qcom: use pm_runtime_resume_and_get() soundwire: intel: use pm_runtime_resume_and_get() soundwire: cadence: use pm_runtime_resume_and_get() soundwire: bus: use pm_runtime_resume_and_get() soundwire: qcom: return error when pm_runtime_get_sync fails soundwire: bus: pm_runtime_request_resume on peripheral attachment soundwire: intel: disable WAKEEN in pm_runtime resume soundwire: intel: prevent pm_runtime resume prior to system suspend soundwire: cadence: recheck device0 attachment after status change dt-bindings: soundwire: qcom: Add bindings for audio clock reset control property soundwire: qcom: Add compatible name for v1.6.0 soundwire: stream: Fix error return code in do_bank_switch() soundwire: qcom: fix an error message in swrm_wait_for_frame_gen_enabled()
-
Greg Kroah-Hartman authored
Merge tag 'phy-for-5.19' of git://git.kernel.org/pub/scm/linux/kernel/git/phy/linux-phy into char-work-next Vinod writes: phy-for-5.19 - New support: - LVDS configuration support and implementation in fsl driver - Qualcomm UFS phy support for SM6350 and USB PHY for SDX65 - Allwinner D-PHY Rx mode support - Yamilfy Mixel mipi-dsi-phy - Updates: - Documentation for phy ops order - Can transceiver mux support - Qualcomm QMP phy updates - Uniphier phy updates * tag 'phy-for-5.19' of git://git.kernel.org/pub/scm/linux/kernel/git/phy/linux-phy: (40 commits) phy: qcom-qmp: rename error labels phy: qcom-qmp: fix pipe-clock imbalance on power-on failure phy: qcom-qmp: switch to explicit reset helpers phy: qcom-qmp: fix reset-controller leak on probe errors phy: qcom-qmp: fix struct clk leak on probe errors dt-bindings: phy: renesas,usb2-phy: Document RZ/G2UL phy bindings dt-bindings: phy: marvell,armada-3700-utmi-host-phy: Fix incorrect compatible in example phy: qcom-qmp: fix phy-descriptor kernel-doc typo phy: rockchip-inno-usb2: Clean up some inconsistent indenting phy: freescale: imx8m-pcie: Handle IMX8_PCIE_REFCLK_PAD_UNUSED phy: core: Warn when phy_power_on is called before phy_init phy: core: Update documentation syntax phy: core: Add documentation of phy operation order phy: rockchip-inno-usb2: Handle ID IRQ phy: rockchip-inno-usb2: Handle bvalid falling phy: rockchip-inno-usb2: Support multi-bit mask properties phy: rockchip-inno-usb2: Do not lock in bvalid IRQ handler phy: rockchip-inno-usb2: Do not check bvalid twice phy: rockchip-inno-usb2: Fix muxed interrupt support phy: allwinner: phy-sun6i-mipi-dphy: Support D-PHY Rx mode for MIPI CSI-2 ...
-
Greg Kroah-Hartman authored
Merge tag 'mhi-for-v5.19' of git://git.kernel.org/pub/scm/linux/kernel/git/mani/mhi into char-work-next Manivannan writes: MHI changes for v5.19 MHI Host -------- Support for new modems: - Foxconn Cinterion MV32-WA/MV32-WB based on SDX62/SDX65 - Telit FN980 v1 based on SDX55 - Telit FN990 based on SDX65 - Foxconn T99W373/T99W368 based on SDX62/SDX65 Core changes: - During the recycle of event ring elements, compute the ctxt_wp based on the local cached value instead of reading from shared memory. This is to prevent the possible corruption of the ctxt_wp as some of the endpoint devices could modify the value in shared memory. - Add sysfs support for resetting the endpoint based on the MHI spec. The MHI spec allows the host to hard reset the device in the case of an unrecoverable error and all other reset mechanisms have failed. - During MHI shutdown, wait for the endpoint device to enter the ready state post reset before proceeding. This is to avoid a possible race where host would remove the interrupt handler and device will send ready state interrupt, resulting in IOMMU fault. - Bail out updating the MHI register if the read has failed during read/modify/write. - Use mhi_write_reg() instead of mhi_write_reg_field() for writing the whole register fields in mhi_init_mmio(). MAINTAINERS change: - Since Qualcomm has moved the email domain for its employess from codeaurora domain to quicinc, update the same for Hemant. * tag 'mhi-for-v5.19' of git://git.kernel.org/pub/scm/linux/kernel/git/mani/mhi: (29 commits) bus: mhi: host: Add support for Foxconn T99W373 and T99W368 bus: mhi: host: pci_generic: add Telit FN990 bus: mhi: host: pci_generic: add Telit FN980 v1 hardware revision bus: mhi: host: Add support for Cinterion MV32-WA/MV32-WB bus: mhi: host: Optimize and update MMIO register write method bus: mhi: host: Bail on writing register fields if read fails bus: mhi: host: Wait for ready state after reset bus: mhi: host: Add soc_reset sysfs bus: mhi: host: pci_generic: Sort mhi_pci_id_table based on the PID bus: mhi: host: Use cached values for calculating the shared write pointer MAINTAINERS: Update Hemant's email id bus: mhi: ep: Add uevent support for module autoloading bus: mhi: ep: Add support for suspending and resuming channels bus: mhi: ep: Add support for queueing SKBs to the host bus: mhi: ep: Add support for processing channel rings bus: mhi: ep: Add support for reading from the host bus: mhi: ep: Add support for processing command rings bus: mhi: ep: Add support for handling SYS_ERR condition bus: mhi: ep: Add support for handling MHI_RESET bus: mhi: ep: Add support for powering down the MHI endpoint stack ...
-
Greg Kroah-Hartman authored
Merge tag 'iio-for-5.19a' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio into char-misc-next Jonathan writes: First set of IIO new device support, features and cleanup for 5.19 Usual mixed bag. Stand out this time is Andy Shevchenko's continuing effort to move drivers over the generic firmware interfaces. Device support * sprd,sc2720 - upm9620 binding addition. - Refactor and support for sc2720, sc2721 and sc2730. * ti,ads1015 - Refactor driver and add support for TLA2024. Device support (IDs only) * invensense,mpu6050 - Add ID for ICM-20608-D. * st,accel: - Add ID for lis302dl. * st,lsm6dsx - Add support for ASM330LHHX (can fallback to LSM6DSR.) Features * convert drivers to device properties - IIO core - adi,ad7266 - adi,adis16480 - adi,adxl355 - bosch,bmi160 - domintech,dmard06 - fsl,fxas21002c - invensense,mpu3050 - linear,ltc2983 - linear,ltc2632 - maxbotix,mb1232 - maxim,max31856 - maxim,max31865 - multiplexer - ping - rescale - taos,tsl2772 * core - Add runtime check on whether realbits fit in storagebits for each channel. * adi,ad_sigma_delta - Add sequencer support and relevant update_scan_mode callbacks for adi,ad7192 and adi,ad7124. Cleanup and minor fixes * MAINTAINERS - Update Lorenzo Bianconi's email address for IIO drivers. - Add entry for ad3552r and update maintainer in dt-binding doc. * tree-wide - Replace strtobool() with kstrtobool(). - Drop false OF dependencies. * core - Tidy up and document IIO modes. - Take iio_buffer_enabled() out of header allowing current_mode to be moved to the opaque structure. - As all kfifo buffers use the same mode value, drop that parameter and set it unconditionally. - White space fixes and similar. - Drop use of list iterator variable for list_for_each_entry_continue_reverse and use list_prepare_entry to restart. * sysfs-trigger - Replace use of 'found' variable with dedicate list iterator variable. * adi,ad7124 - Drop misleading shift. * adi,ad2s1210 - Remove redundant local variable assignment. * adi,adis16480 - Use local device pointer to reduce repetition. - Improve handling of clocks. * domintech,dmard09 - White space. * dummy driver - Improve error handling. * fsl,mma8452 - Add missing documentation of name element. * invensense,mpu3050 - Stop remove() returning non 0. * kionix,kxsd9 - White space. * linear,ltc2688 - Use local variable for struct device. - Combine of_node_put() error handling paths. * linear,ltc2983 - Avoid use of constants in messages where a define is available. * microchip,mcp4131 - Fix compatible in dt example. * pni,rm3100 - Stop directly accessing iio_dev->current_mode just to find out if the buffer is enabled. * renesas,rzg2l - Relax kconfig constraint to include newer devices. * sprd,sc27xx - Fix wrong scaling mask. - Improve the calibration values. * samsung,ssp - Replace a 'found' variable in favor of an explicit value that was found. * sensortek,stk3xx - Add proximity-near-level binding and driver support. * st,st_sensors: - Drop unused accel_type enum. - Return early in *_write_raw() - Drop unnecessary locking in _avail functions. - Add local lock to protect odr against concurrent updates allowing mlock to no longer be used outside of the core. - Use iio_device_claim_direct_mode() rather than racy checking of the current mode. * st,stmpe-adc - Fix checks on wait_for_completion_timeout(). - Allow use of of_device_id for matching. * st,stm32-dfsdm - Stop accessing iio_dev->current_mode to find out if the buffer is enabled (so we can hide that variable in the opaque structure) * st,vl53l0x - Fix checks on wait_for_completion_timeout. * ti,ads1015 - Add missing ID for ti,ads1115 in binding doc. - Convert from repeated chip ID look up to selecting static const data. - Switch to read_avail() callback. * ti,ads8688 - Use of_device_id for driver matching. * ti,palmas-adc - Drop a warning on minor calibration mismatch leading to slightly negative values after applying the calibration. * tag 'iio-for-5.19a' of https://git.kernel.org/pub/scm/linux/kernel/git/jic23/iio: (95 commits) iio: ti-ads8688: use of_device_id for OF matching iio: stmpe-adc: use of_device_id for OF matching dt-bindings: iio: Fix incorrect compatible strings in examples iio: gyro: mpu3050: Make mpu3050_common_remove() return void iio: dac: ltc2632: Make use of device properties iio: temperature: max31865: Make use of device properties iio: proximity: mb1232: Switch to use fwnode_irq_get() iio: imu: adis16480: Improve getting the optional clocks iio: imu: adis16480: Use temporary variable for struct device iio: imu: adis16480: Make use of device properties staging: iio: ad2s1210: remove redundant assignment to variable negative iio: adc: sc27xx: add support for PMIC sc2730 iio: adc: sc27xx: add support for PMIC sc2720 and sc2721 iio: adc: sc27xx: refactor some functions for support more PMiCs iio: adc: sc27xx: structure adjustment and optimization iio: adc: sc27xx: Fine tune the scale calibration values iio: adc: sc27xx: fix read big scale voltage not right dt-bindings:iio:adc: add sprd,ump9620-adc dt-binding iio: proximity: stk3310: Export near level property for proximity sensor dt-bindings: iio: light: stk33xx: Add proximity-near-level ...
-
- 10 May, 2022 3 commits
-
-
Slark Xiao authored
Product's enumeration align with previous Foxconn SDX55, so T99W373(SDX62)/T99W368(SDX65) would use the same config as Foxconn SDX55. Remove fw and edl for this new commit. Signed-off-by:
Slark Xiao <slark_xiao@163.com> Reviewed-by:
Manivannan Sadhasivam <mani@kernel.org> Reviewed-by:
Loic Poulain <loic.poulain@linaro.org> Link: https://lore.kernel.org/r/20220503024349.4486-1-slark_xiao@163.comSigned-off-by:
Manivannan Sadhasivam <manivannan.sadhasivam@linaro.org>
-
Daniele Palmas authored
Add Telit FN990: 01:00.0 Unassigned class [ff00]: Qualcomm Device 0308 Subsystem: Device 1c5d:2010 Signed-off-by:Daniele Palmas <dnlplm@gmail.com> Reviewed-by:
-
Daniele Palmas authored
Add Telit FN980 v1 hardware revision: 01:00.0 Unassigned class [ff00]: Qualcomm Device [17cb:0306] Subsystem: Device [1c5d:2000] Signed-off-by:
-
- 09 May, 2022 33 commits
-
-
Sean Anderson authored
This adds support for Trust Architecture (TA) 2.1 devices to the SFP driver. There are few differences between TA 2.1 and TA 3.0, especially for read-only support, so just re-use the existing data. Signed-off-by:
Sean Anderson <sean.anderson@seco.com> Signed-off-by:
Srinivas Kandagatla <srinivas.kandagatla@linaro.org> Link: https://lore.kernel.org/r/20220429162701.2222-17-srinivas.kandagatla@linaro.orgSigned-off-by:
Greg Kroah-Hartman <gregkh@linuxfoundation.org>
-
Sean Anderson authored
This converts the SFP driver to use regmap. This will allow easily supporting devices with different endians. We disallow byte-level access, as regmap_bulk_read doesn't support it (and it's unclear what the correct result would be when we have an endianness difference). Signed-off-by:
-
Sean Anderson authored
Trust Architecture (TA) 2.1 devices include the LS1012A, LS1021A, LS1043A, and LS1046A. The SFP device on TA 2.1 devices is very similar to the SFP on TA 3.0 devices. The primary difference is a few fields in the control register. Add a compatible string. Signed-off-by:
-
Sean Anderson authored
The TA_PROG_SFP supply must be enabled to program the fuses, and disabled to read the fuses (such as at power-on-reset). On many boards, this supply is controlled by a jumper. The user must manually insert or remove it at the appropriate time in the programming process. However, on other boards this supply is controlled by an FPGA or a GPIO. In these cases, the driver can automatically enable and disable it as necessary. Reviewed-by:
Andrew Lunn <andrew@lunn.ch> Signed-off-by:
-
Sean Anderson authored
To program fuses, it is necessary to set the fuse programming time. This is determined based on the value of the platform clock. Add a clock property. Because this property is necessary for programming, it is made mandatory. Since these bindings have not yet been present in a stable release (though they are on track for 5.18), it is not an ABI break to change them in this manner. Signed-off-by:
-
Sean Anderson authored
There is a small grammatical error in the description. Fix it. Reviewed-by:
Michael Walle <michael@walle.cc> Reviewed-by:
-
Minghao Chi authored
Using pm_runtime_resume_and_get is more appropriate for simplifing code Reported-by:
Zeal Robot <zealci@zte.com.cn> Signed-off-by:
Minghao Chi <chi.minghao@zte.com.cn> Signed-off-by:
-
Sven Peter authored
Apple SoCs contain eFuses used to store factory-programmed data such as calibration values for the PCIe or the Type-C PHY. They are organized as 32bit values exposed as MMIO. Signed-off-by:
Sven Peter <sven@svenpeter.dev> Signed-off-by:
-
Sven Peter authored
Apple SoCs come with eFuses used to store factory-programmed data such as calibration settings for the PCIe and Type-C PHY. Reviewed-by:
Krzysztof Kozlowski <krzysztof.kozlowski@canonical.com> Acked-by:
Rob Herring <robh@kernel.org> Signed-off-by:
-
Sven Peter authored
Signed-off-by:
-
Rafał Miłecki authored
DT binding for Broadcom's NVRAM supports specifying NVMEM cells as NVMEM device (provider) subnodes. Look for such subnodes when collecing NVMEM cells. This allows NVMEM consumers to use NVRAM variables. Signed-off-by:
Rafał Miłecki <rafal@milecki.pl> Signed-off-by:
-
Rafał Miłecki authored
Some hardware may have NVMEM cells described in Device Tree using individual nodes. Let drivers pass such nodes to the NVMEM subsystem so they can be later used by NVMEM consumers. Signed-off-by:
-
Carlos Llamas authored
Log readable and specific error messages whenever a transaction failure happens. This will ensure better context is given to regular users about these unique error cases, without having to decode a cryptic log. Acked-by:
Todd Kjos <tkjos@google.com> Acked-by:
Christian Brauner (Microsoft) <brauner@kernel.org> Signed-off-by:
Carlos Llamas <cmllamas@google.com> Link: https://lore.kernel.org/r/20220429235644.697372-6-cmllamas@google.comSigned-off-by:
-
Carlos Llamas authored
Converting binder_debug() and binder_user_error() macros into functions reduces the overall object size by 16936 bytes when cross-compiled with aarch64-linux-gnu-gcc 11.2.0: $ size drivers/android/binder.o.{old,new} text data bss dec hex filename 77935 6168 20264 104367 197af drivers/android/binder.o.old 65551 1616 20264 87431 15587 drivers/android/binder.o.new This is particularly beneficial to functions binder_transaction() and binder_thread_write() which repeatedly use these macros and are both part of the critical path for all binder transactions. $ nm --size vmlinux.{old,new} |grep ' binder_transaction$' 0000000000002f60 t binder_transaction 0000000000002358 t binder_transaction $ nm --size vmlinux.{old,new} |grep binder_thread_write 0000000000001c54 t binder_thread_write 00000000000014a8 t binder_thread_write Acked-by: -
Carlos Llamas authored
Add extended_error to the binderfs feature list, to help userspace determine whether the BINDER_GET_EXTENDED_ERROR ioctl is supported by the binder driver. Reviewed-by:
-
Carlos Llamas authored
Provide a userspace mechanism to pull precise error information upon failed operations. Extending the current error codes returned by the interfaces allows userspace to better determine the course of action. This could be for instance, retrying a failed transaction at a later point and thus offloading the error handling from the driver. Acked-by:
-
Carlos Llamas authored
Make sure we log relevant information about failed transactions such as the target proc/thread, call type and transaction id. These details are particularly important when debugging userspace issues. Acked-by:
-
Matthew Gerlach authored
Add the Device Feature List (DFL) feature id for the High Speed Serial Interface (HSSI) Subsystem to the table of ids supported by the uio_dfl driver. The HSSI Subsystem is a configurable set of IP blocks to be used as part of a Ethernet or PCS/FEC/PMA pipeline. Like the Ethernet group used by the N3000 card, the HSSI Subsystem does not fully implement a network device from a Linux netdev perspective and is controlled and monitored from user space software via the uio interface. The Feature ID table of DFL can be found: https://github.com/OPAE/dfl-feature-idReviewed-by:
Tom Rix <trix@redhat.com> Acked-by:
Xu Yilun <yilun.xu@intel.com> Signed-off-by:
Matthew Gerlach <matthew.gerlach@linux.intel.com> Signed-off-by:
Tianfei Zhang <tianfei.zhang@intel.com> Link: https://lore.kernel.org/r/20220505094129.686535-1-tianfei.zhang@intel.comSigned-off-by:
-
Lad Prabhakar authored
platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static allocation of IRQ resources in DT core code, this causes an issue when using hierarchical interrupt domains using "interrupts" property in the node as this bypasses the hierarchical setup and messes up the irq chaining. In preparation for removal of static setup of IRQ resource from DT core code use platform_get_irq(). Tested-By:
Steev Klimaszewski <steev@kali.org> Reviewed-by:
Bjorn Andersson <bjorn.andersson@linaro.org> Signed-off-by:
Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com> Signed-off-by:
-
Wan Jiabing authored
Fix following coccicheck warning: ./drivers/bus/mhi/host/init.c:89:8-16: WARNING: use scnprintf or sprintf Use sysfs_emit and sysfs_emit_at instead of snprintf. Reviewed-by:
Kees Cook <keescook@chromium.org> Reviewed-by:
Wan Jiabing <wanjiabing@vivo.com> Link: https://lore.kernel.org/r/20220426125902.681258-1-wanjiabing@vivo.comSigned-off-by:
-
Kees Cook authored
The masking for PAC bits wasn't handling 32-bit architectures correctly. Replace the u64 cast with uintptr_t. Link: https://lore.kernel.org/lkml/CAMuHMdVz-J-1ZQ08u0bsQihDkcRmEPrtX5B_oRJ+Ns5jrasnUw@mail.gmail.com Fixes: 2e53b877 ("lkdtm: Add CFI_BACKWARD to test ROP mitigations") Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: Arnd Bergmann <arnd@arndb.de> Reported-by:
kernel test robot <lkp@intel.com> Reported-by:
Geert Uytterhoeven <geert@linux-m68k.org> Tested-by:
Randy Dunlap <rdunlap@infradead.org> Signed-off-by:
-
Shuah Khan authored
rtsx_usb_probe() doesn't call usb_set_intfdata() to null out the interface pointer when probe fails. This leaves a stale pointer. Noticed the missing usb_set_intfdata() while debugging an unrelated invalid DMA mapping problem. Fix it with a call to usb_set_intfdata(..., NULL). Signed-off-by:
Shuah Khan <skhan@linuxfoundation.org> Link: https://lore.kernel.org/r/20220429210913.46804-1-skhan@linuxfoundation.orgSigned-off-by:
-
Jiapeng Chong authored
Clean the following coccicheck warning: ./drivers/misc/altera-stapl/altera.c:955:51-52: WARNING opportunity for swap(). Reported-by:
Abaci Robot <abaci@linux.alibaba.com> Signed-off-by:
Jiapeng Chong <jiapeng.chong@linux.alibaba.com> Link: https://lore.kernel.org/r/20220505081539.91575-1-jiapeng.chong@linux.alibaba.comSigned-off-by:
-
Ricky WU authored
move rts5261_fetch_vendor_settings() to rts5261_init_from_hw() make sure it be called from S3 or D3 add more register setting when efuse is set read efuse setting to register on init flow Signed-off-by:
Ricky Wu <Ricky_wu@realtek.com> Link: https://lore.kernel.org/r/18101ecb0f0749ccb9f564eda171ba40@realtek.comSigned-off-by:
-
Srinivas Kandagatla authored
Currently timeout for autoenumeration during probe and bus reset is set to 2 secs which is really a big value. This can have an adverse effect on boot time if the slave device is not ready/reset. This was the case with wcd938x which was not reset yet but we spent 2 secs waiting in the soundwire controller probe. Reduce this time to 1/10 of Hz which should be good enough time to finish autoenumeration if any slaves are available on the bus. Reported-by:
Srinivasa Rao Mandadapu <quic_srivasam@quicinc.com> Signed-off-by:
Vinod Koul <vkoul@kernel.org>
-
Pierre-Louis Bossart authored
Use pm_runtime_resume_and_get() to replace the pm_runtime_get_sync() and pm_runtime_put_noidle() pattern. No functional changes. Signed-off-by:
Pierre-Louis Bossart <pierre-louis.bossart@linux.intel.com> Reviewed-by:
Péter Ujfalusi <peter.ujfalusi@linux.intel.com> Signed-off-by:
Bard Liao <yung-chuan.liao@linux.intel.com> Reviewed-by:
-
Pierre-Louis Bossart authored
Use pm_runtime_resume_and_get() to replace the pm_runtime_get_sync() and pm_runtime_put_noidle() pattern. No functional changes. Signed-off-by:
-
Pierre-Louis Bossart authored
Use pm_runtime_resume_and_get() to replace the pm_runtime_get_sync() and pm_runtime_put_noidle() pattern. No functional changes. Signed-off-by:
-
Pierre-Louis Bossart authored
Use pm_runtime_resume_and_get() to replace the pm_runtime_get_sync() and pm_runtime_put_noidle() pattern. No functional changes. Signed-off-by:
-
Pierre-Louis Bossart authored
For some reason there's a missing error return in two places. Fixes: 74e79da9 ("soundwire: qcom: add runtime pm support") Fixes: 04d46a7b ("soundwire: qcom: add in-band wake up interrupt support") Signed-off-by:
-
Pierre-Louis Bossart authored
In typical use cases, the peripheral becomes pm_runtime active as a result of the ALSA/ASoC framework starting up a DAI. The parent/child hierarchy guarantees that the manager device will be fully resumed beforehand. There is however a corner case where the manager device may become pm_runtime active, but without ALSA/ASoC requesting any functionality from the peripherals. In this case, the hardware peripheral device will report as ATTACHED and its initialization routine will be executed. If this initialization routine initiates any sort of deferred processing, there is a possibility that the manager could suspend without the peripheral suspend sequence being invoked: from the pm_runtime framework perspective, the peripheral is *already* suspended. To avoid such disconnects between hardware state and pm_runtime state, this patch adds an asynchronous pm_request_resume() upon successful attach/initialization which will result in the proper resume/suspend sequence to be followed on the peripheral side. BugLink: https://github.com/thesofproject/linux/issues/3459Signed-off-by:
Ranjani Sridharan <ranjani.sridharan@linux.intel.com> Reviewed-by:
Rander Wang <rander.wang@intel.com> Signed-off-by:
-
Pierre-Louis Bossart authored
When the manager device is pm_runtime resumed, we see a series of spurious wakes and attempts to resume the same device: soundwire_intel.link.0: intel_resume_runtime: start soundwire_intel.link.0: intel_link_power_up: powering up all links soundwire_intel.link.0: intel_link_power_up: first link up, programming SYNCPRD soundwire_intel.link.0: intel_shim_wake: WAKEEN disabled for link 0 soundwire_intel.link.0: intel_link_process_wakeen_event: pm_request_resume start soundwire_intel.link.0: intel_link_process_wakeen_event: pm_request_resume done soundwire_intel.link.0: intel_shim_wake: WAKEEN disabled for link 0 soundwire_intel.link.0: intel_link_process_wakeen_event: pm_request_resume start soundwire_intel.link.0: intel_link_process_wakeen_event: pm_request_resume done This sequence does not break anything but is totally unnecessary. Currently the wakes are only disabled after the peripheral generates a wake, e.g. for jack detection. If the resume is initiated by the host drivers as a result of userspace actions (play/record typically), we need to disable wake detection as well. Doing so prevents the spurious wakes and calls to pm_request_resume(). Signed-off-by:
-
Pierre-Louis Bossart authored
commit e38f9ff6 ("ACPI: scan: Do not add device IDs from _CID if _HID is not valid") exposes a race condition on a TGL RVP device leading to a timeout. The detailed analysis shows the RT711 codec driver scheduling a jack detection workqueue while attaching during a spurious pm_runtime resume, and the work function happens to be scheduled after the manager device is suspended. The direct link between this ACPI patch and a spurious pm_runtime resume is not obvious; the most likely explanation is that a change in the ACPI device linked list management modifies the order in which the pm_runtime device status is checked and exposes a race condition that was probably present for a very long time, but was not identified. We already have a check in the .prepare stage, where we will resume to full power from specific clock-stop modes. In all other cases, we don't need to resume to full power by default. Adding the SMART_SUSPEND flag prevents the spurious resume from happening. BugLink: https://github.com/thesofproject/linux/issues/3459 Fixes: 029bfd1c ("soundwire: intel: conditionally exit clock stop mode on system suspend") Signed-off-by:
-
