The mvebu-comphy driver does not currently know how to pass correct
lane-count to ATF while configuring the serdes lanes.
This causes the system to hard reset during reconfiguration, if a pci
card is present and has established a link during bootloader.
Remove the comphy handles from the respective pci nodes to avoid runtime
reconfiguration, relying solely on bootloader configuration - while
avoiding the hard reset.
When bootloader has configured the lanes correctly, the pci ports are
functional under Linux.
This issue may be addressed in the comphy driver at a future point.
Fixes: e9ff907f40 ("arm64: dts: add description for solidrun cn9132 cex7 module and clearfog board")
Cc: stable@vger.kernel.org
Signed-off-by: Josua Mayer <josua@solid-run.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@bootlin.com>
Similar to MacchiatoBIN the high-speed modes are unstable on the CN9132
CEX-7 module, leading to failed transactions under normal use.
Disable all high-speed modes including UHS.
Additionally add no-sdio and non-removable properties as appropriate for
eMMC.
Fixes: e9ff907f40 ("arm64: dts: add description for solidrun cn9132 cex7 module and clearfog board")
Cc: stable@vger.kernel.org
Signed-off-by: Josua Mayer <josua@solid-run.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@bootlin.com>
Commit "arm64: dts: marvell: only enable complete sata nodes" changed
armada-cp11x.dtsi disabling all sata ports status by default.
The author missed some dts which relied on the dtsi enabling all ports,
and just disabled unused ones instead.
Update dts for SolidRun cn913x based boards to enable the available
ports, rather than disabling the unvavailable one.
Further according to dt bindings the serdes phys are to be specified in
the port node, not the controller node.
Move those phys properties accordingly in clearfog base/pro/solidwan.
Fixes: 30023876ae ("arm64: dts: marvell: only enable complete sata nodes")
Cc: stable@vger.kernel.org
Signed-off-by: Josua Mayer <josua@solid-run.com>
Signed-off-by: Gregory CLEMENT <gregory.clement@bootlin.com>
Introduce a new device tree flag to cap the maximum High-Speed (HS)
mode frequency for SD cards, accommodating board-specific
electrical limitations which cannot support the default 50Mhz HS
frequency and others.
Signed-off-by: Sarthak Garg <quic_sartgarg@quicinc.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Introduce a new optional device tree property max-sd-hs-hz to
limit the maximum frequency (in Hz) used for SD cards operating in
High-Speed (HS) mode due to any board electrical limitations.
Signed-off-by: Sarthak Garg <quic_sartgarg@quicinc.com>
Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
For Qualcomm SoCs which needs level shifter for SD card, extra delay is
seen on receiver data path.
To compensate this delay enable tuning for SDR50 mode for targets which
has level shifter. SDHCI_SDR50_NEEDS_TUNING caps will be set for targets
with level shifter on Qualcomm SOC's.
Signed-off-by: Sarthak Garg <quic_sartgarg@quicinc.com>
Acked-by: Adrian Hunter <adrian.hunter@intel.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Add the MSM SDHCI compatible name to support both eMMC and SD card for
Lemans, which uses 'sa8775p' as the fallback SoC. Ensure the new
compatible string matches existing Lemans-compatible formats without
introducing a new naming convention.
The SDHCI controller on Lemans is based on MSM SDHCI v5 IP. Hence,
document the compatible with "qcom,sdhci-msm-v5" as the fallback.
Signed-off-by: Monish Chunara <quic_mchunara@quicinc.com>
Acked-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Wasim Nazir <wasim.nazir@oss.qualcomm.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Merge the pmdomain fixes for v6.17-rc[n] into the next branch, to allow
them to get tested together with the new changes that are targeted for
v6.18.
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
At the moment the driver sets the power state of all the PDs it creates
to off, regardless of the actual HW state. This has two drawbacks:
1) The kernel cannot disable unused PDs automatically for power saving,
as it thinks they are off already
2) A more specific case (but perhaps applicable to other scenarios
also): bootloader enabled splash-screen cannot be kept on the screen.
The issue in 2) is that the driver framework automatically enables the
device's PD before calling probe() and disables it after the probe().
This means that when the display subsystem (DSS) driver probes, but e.g.
fails due to deferred probing, the DSS PD gets turned off and the driver
cannot do anything to affect that.
Solving the 2) requires more changes to actually keep the PD on during
the boot, but a prerequisite for it is to have the correct power state
for the PD.
The downside with this patch is that it takes time to call the 'is_on'
op, and we need to call it for each PD. In my tests with AM62 SK, using
defconfig, I see an increase from ~3.5ms to ~7ms. However, the added
feature is valuable, so in my opinion it's worth it.
The performance could probably be improved with a new firmware API which
returns the power states of all the PDs.
There's also a related HW issue at play here: if the DSS IP is enabled
and active, and its PD is turned off without first disabling the DSS
display outputs, the DSS IP will hang and causes the kernel to halt if
and when the DSS driver accesses the DSS registers the next time.
With the current upstream kernel, with this patch applied, this means
that if the bootloader enables the display, and the DSS driver is
compiled as a module, the kernel will at some point disable unused PDs,
including the DSS PD. When the DSS module is later loaded, it will hang
the kernel.
The same issue is already there, even without this patch, as the DSS
driver may hit deferred probing, which causes the PD to be turned off,
and leading to kernel halt when the DSS driver is probed again. This
issue has been made quite rare with some arrangements in the DSS
driver's probe, but it's still there.
With recent change from Ulf (e.g. commit 13a4b7fb62 ("pmdomain: core:
Leave powered-on genpds on until late_initcall_sync")), the sync state
mechanism comes to rescue. It will keep the power domains enabled, until
the drivers have probed, or the sync-state is triggered via some other
mechanism (e.g. manually by the boot scripts).
Reviewed-by: Kevin Hilman <khilman@baylibre.com>
Tested-by: Kevin Hilman <khilman@baylibre.com>
Signed-off-by: Tomi Valkeinen <tomi.valkeinen@ideasonboard.com>
Signed-off-by: Ulf Hansson <ulf.hansson@linaro.org>
Guan-Yu Lin <guanyulin@google.com> says:
Wesley Cheng and Mathias Nyman's USB offload design enables a
co-processor to handle some USB transfers, potentially allowing the
system to sleep (suspend-to-RAM) and save power. However, Linux's System
Sleep model halts the USB host controller when the main system isn't
managing any USB transfers. To address this, the proposal modifies the
system to recognize offloaded USB transfers and manage power
accordingly. This way, offloaded USB transfers could still happen during
system sleep (Suspend-to-RAM).
This involves two key steps:
1. Transfer Status Tracking: Propose offload_usage and corresponding
apis drivers could track USB transfers on the co-processor, ensuring
the system is aware of any ongoing activity.
2. Power Management Adjustment: Modifications to the USB driver stack
(xhci host controller driver, and USB device drivers) allow the
system to sleep (Suspend-to-RAM) without disrupting co-processor
managed USB transfers. This involves adding conditional checks to
bypass some power management operations in the System Sleep model.
Link: https://lore.kernel.org/r/20250911142051.90822-1-guanyulin@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Sharing a USB controller with another entity via xhci-sideband driver
creates power management complexities. To prevent the USB controller
from being inadvertently deactivated while in use by the other entity, a
usage-count based mechanism is implemented. This allows the system to
manage power effectively, ensuring the controller remains available
whenever needed.
In order to maintain full functionality of an offloaded USB devices,
several changes are made within the suspend flow of such devices:
- skip usb_suspend_device() so that the port/hub are still active for
USB transfers via offloaded path.
- not suspending the endpoints which are used by USB interfaces marked
with needs_remote_wakeup. Namely, skip usb_suspend_interface() and
usb_hcd_flush_endpoint() on associated USB interfaces. This reserves a
pending interrupt urb during system suspend for handling the interrupt
transfer, which is necessary since remote wakeup doesn't apply in the
offloaded USB devices when controller is still active.
- not flushing the endpoints of actively offloaded USB devices. Given
that the USB devices is used by another entity, unilaterally flush the
endpoint might lead to unexpected behavior on another entity.
- not suspending the xhci controller. This is done by skipping the
suspend/resume callbacks in the xhci platform driver.
Signed-off-by: Guan-Yu Lin <guanyulin@google.com>
Link: https://lore.kernel.org/r/20250911142051.90822-5-guanyulin@google.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Link: https://lore.kernel.org/r/20250911142051.90822-5-guanyulin@google.com
Add a driver for the Intel USBIO USB IO-expander used by the MIPI cameras
on various new (Meteor Lake and later) Intel laptops.
This is an USB bridge driver which adds auxbus child devices for the GPIO,
I2C and SPI functions of the USBIO chip and which exports IO-functions for
the drivers for the auxbus child devices to communicate with the USBIO
device's firmware.
Co-developed-by: Hans de Goede <hansg@kernel.org>
Signed-off-by: Hans de Goede <hansg@kernel.org>
Signed-off-by: Israel Cepeda <israel.a.cepeda.lopez@intel.com>
Link: https://lore.kernel.org/r/20250911181343.77398-2-hansg@kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
pinctrl: renesas: Updates for v6.18 (take two)
- Improve suspend/resume support on RZ/G2L family SoCs,
- Miscellaneous fixes and improvements.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
On Qualcomm DWC3 dual-role controllers, the conndone/disconnect events in
device mode are generated by controller when software writes to QSCRATCH
registers in Qualcomm Glue layer rather than the vbus line being routed to
dwc3 core IP for it to recognize and generate these events.
UTMI_OTG_VBUS_VALID bit of QSCRATCH_HS_PHY_CTRL register needs to be set
to generate a connection done event and to be cleared for the controller to
generate a disconnect event during cable removal. When the disconnect is
not generated upon cable removal, the "connected" flag of dwc3 is left
marked as "true" and it blocks suspend routines and for that to happen upon
cable removal, the cable disconnect notification coming in via set_role
call need to be provided to the Qualcomm glue layer as well.
Currently, the way DWC3 core and Qualcomm legacy glue driver are designed,
there is no mechanism through which the DWC3 core can notify the Qualcomm
glue layer of any role changes which it receives via role switch. To
register these glue callbacks at probe time, for enabling core to notify
glue layer, the legacy Qualcomm driver has no way to find out when the
child driver probe was successful since it does not check for the same
during of_platform_populate.
Hence implement the following glue callbacks for flattened Qualcomm glue
driver:
1. set_role: To pass role switching information from drd layer to glue.
This information is needed to identify NONE/DEVICE mode switch and modify
QSCRATCH to generate connect-done event on device mode entry and disconnect
event on cable removal in device mode.
2. run_stop: When booting up in device mode, if autouspend is enabled and
userspace doesn't write UDC on boot, controller enters autosuspend. After
this, if the userspace writes to UDC in the future, run_stop notifier is
required to enable UTMI_OTG_VBUS_VALID of QSCRATCH so that connect done
event is generated after run_stop(1) is done to finish enumeration.
Acked-by: Thinh Nguyen <Thinh.Nguyen@synopsys.com>
Signed-off-by: Krishna Kurapati <krishna.kurapati@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250907181412.2174616-3-krishna.kurapati@oss.qualcomm.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In certain situations like role switching, the glue layers need to be
informed of these events, so that they can take any necessary action.
But in non-flattened implementations, the glue drivers have no data on
when the core driver probe was successful post invoking of_platform_
populate. Now that the core driver supports flattened implementations
as well, introduce vendor callbacks that can be passed on from glue to
core before invoking dwc3_core_probe.
Introduce callbacks to notify glue layer of role_switch and run_stop
changes. These can be used by flattened implementation of Qualcomm
glue layer to generate connect/disconnect events in controller during
cable connect and run stop modifications by udc in device mode.
Acked-by: Thinh Nguyen <Thinh.Nguyen@synopsys.com>
Signed-off-by: Krishna Kurapati <krishna.kurapati@oss.qualcomm.com>
Link: https://lore.kernel.org/r/20250907181412.2174616-2-krishna.kurapati@oss.qualcomm.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
When the system is suspended, USB hot-plugging/unplugging can trigger
wake events of the Tegra USB host controller.
Enable support for USB wake-up events by parsing device-tree to see if
the interrupts for the wake-up events are present and if so configure
those interrupts. Note that if wake-up events are not present, still
allow the USB host controller to probe as normal.
Signed-off-by: Haotien Hsu <haotienh@nvidia.com>
Link: https://lore.kernel.org/r/20250811074558.1062048-5-haotienh@nvidia.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
The DWC2 IP can be properly integrated in a SoC to work at high-speed
USB speed but some board issues, EMC constraints or any other reasons
can lead to the need to limit this USB speed at board level.
The device-tree 'maximum-speed' property already exists for this purpose
but is not handled by the DWC2 driver.
Fill this lack adding support for 'maximum-speed' property and so allow
to limit the USB speed in device-tree (board description).
Signed-off-by: Herve Codina <herve.codina@bootlin.com>
Link: https://lore.kernel.org/r/20250910160730.585303-1-herve.codina@bootlin.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add wakeup support for tcpci. If the user enables the wakeup file, call
enable_irq_wake() during system suspend and disable_irq_wake() during
system resume. Since this driver supports shared interrupts, mask the
chip interrupt by default when wakeup is disabled to avoid affecting
other IRQ users.
Reviewed-by: Frank Li <Frank.Li@nxp.com>
Reviewed-by: Badhri Jagan Sridharan <badhri@google.com>
Signed-off-by: Xu Yang <xu.yang_2@nxp.com>
Link: https://lore.kernel.org/r/20250910105042.695146-1-xu.yang_2@nxp.com
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Samsung pinctrl drivers changes for v6.18
1. Add pin controller drivers for new Axis ARTPEC-8 SoC. The SoC shares
all main blocks, including the pin controller, with Samsung SoC, so
same drivers and bindings are used.
2. Drop remaining support for Samsung S3C2410 SoC pin controllers. The
actual SoC support was removed in January 2023, so this is just
remaining cleanup.
Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
While the GCC and Clang compilers already define __ASSEMBLER__
automatically when compiling assembly code, __ASSEMBLY__ is a
macro that only gets defined by the Makefiles in the kernel.
This can be very confusing when switching between userspace
and kernelspace coding, or when dealing with uapi headers that
rather should use __ASSEMBLER__ instead. So let's standardize on
the __ASSEMBLER__ macro that is provided by the compilers now.
This is a completely mechanical patch (done with a simple "sed -i"
statement).
Cc: Michal Simek <monstr@monstr.eu>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Link: https://lore.kernel.org/r/20250314071013.1575167-19-thuth@redhat.com
Signed-off-by: Michal Simek <michal.simek@amd.com>
__ASSEMBLY__ is only defined by the Makefile of the kernel, so
this is not really useful for uapi headers (unless the userspace
Makefile defines it, too). Let's switch to __ASSEMBLER__ which
gets set automatically by the compiler when compiling assembly
code.
This is a completely mechanical patch (done with a simple "sed -i"
statement).
Cc: Michal Simek <monstr@monstr.eu>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Link: https://lore.kernel.org/r/20250314071013.1575167-18-thuth@redhat.com
Signed-off-by: Michal Simek <michal.simek@amd.com>
Add interconnect dt-bindings and driver support for
Qualcomm's next gen compute SoC - Glymur.
* icc-glymur
dt-bindings: interconnect: document the RPMh Network-On-Chip interconnect in Glymur SoC
interconnect: qcom: icc-rpmh: increase MAX_PORTS to support four QoS ports
interconnect: qcom: add glymur interconnect provider driver
Link: https://lore.kernel.org/r/20250814-glymur-icc-v2-0-596cca6b6015@oss.qualcomm.com
Signed-off-by: Georgi Djakov <djakov@kernel.org>
The mhi_ep_read_channel function incorrectly assumes the End of Transfer
(EOT) bit is present for each packet in a chained transactions, causing
it to advance mhi_chan->rd_offset beyond wr_offset during host-to-device
transfers when EOT has not yet arrived. This leads to access of unmapped
host memory, causing IOMMU faults and processing of stale TREs.
Modify the loop condition to ensure mhi_queue is not empty, allowing the
function to process only valid TREs up to the current write pointer to
prevent premature reads and ensure safe traversal of chained TREs.
Due to this change, buf_left needs to be removed from the while loop
condition to avoid exiting prematurely before reading the ring completely,
and also remove write_offset since it will always be zero because the new
cache buffer is allocated every time.
Fixes: 5301258899 ("bus: mhi: ep: Add support for reading from the host")
Co-developed-by: Akhil Vinod <akhil.vinod@oss.qualcomm.com>
Signed-off-by: Akhil Vinod <akhil.vinod@oss.qualcomm.com>
Signed-off-by: Sumit Kumar <sumit.kumar@oss.qualcomm.com>
[mani: reworded description slightly]
Signed-off-by: Manivannan Sadhasivam <manivannan.sadhasivam@oss.qualcomm.com>
Reviewed-by: Krishna Chaitanya Chundru <krishna.chundru@oss.qualcomm.com>
Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20250910-final_chained-v3-1-ec77c9d88ace@oss.qualcomm.com
We already have dedicated helper macros for printing GT-oriented
messages but we don't have any to print messages that are tile
oriented and we wrongly try to use plain drm or GT-oriented ones.
Add tile-oriented printk messages and to provide similar coverage
as we have with xe_assert() macros. Also add set of simple macros
for the top level xe_device, which we could easily tweak to include
extra device specific info if needed.
Typical output of our printk macros will look like:
[drm] this is xe_WARN()
[drm] *ERROR* this is xe_err()
[drm] *ERROR* this is xe_err_printer()
[drm] this is xe_info()
[drm] this is xe_info_printer()
[drm:printk_demo.cold] this is xe_dbg()
[drm:printk_demo.cold] this is xe_dbg_printer()
[drm] Tile0: this is xe_tile_WARN()
[drm] *ERROR* Tile0: this is xe_tile_err()
[drm] *ERROR* Tile0: this is xe_tile_err_printer()
[drm] Tile0: this is xe_tile_info()
[drm] Tile0: this is xe_tile_info_printer()
[drm:printk_demo.cold] Tile0: this is xe_tile_dbg()
[drm:printk_demo.cold] Tile0: this is xe_tile_dbg_printer()
[drm] Tile0: GT0: this is xe_gt_WARN()
[drm] *ERROR* Tile0: GT0: this is xe_gt_err()
[drm] *ERROR* Tile0: GT0: this is xe_gt_err_printer()
[drm] Tile0: GT0: this is xe_gt_info()
[drm] Tile0: GT0: this is xe_gt_info_printer()
[drm:printk_demo.cold] Tile0: GT0: this is xe_gt_dbg()
[drm:printk_demo.cold] Tile0: GT0: this is xe_gt_dbg_printer()
Signed-off-by: Michal Wajdeczko <michal.wajdeczko@intel.com>
Cc: Lucas De Marchi <lucas.demarchi@intel.com>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
Reviewed-by: Lucas De Marchi <lucas.demarchi@intel.com>
Link: https://lore.kernel.org/r/20250909165941.31730-5-michal.wajdeczko@intel.com
The ITS mitigation includes both indirect branch thunks and return
thunks. Both are currently placed in .text..__x86.return_thunk, which is
appropriate for the latter but not the former.
For consistency with other mitigations, move the indirect branch thunks to
.text..__x86.indirect_thunk.
Signed-off-by: Josh Poimboeuf <jpoimboe@kernel.org>
Signed-off-by: Borislav Petkov (AMD) <bp@alien8.de>
Reviewed-by: Pawan Gupta <pawan.kumar.gupta@linux.intel.com>
In gdlm_put_lock(), there is a small window of time in which the
DFL_UNMOUNT flag has been set but the lockspace hasn't been released,
yet. In that window, dlm may still call gdlm_ast() and gdlm_bast().
To prevent it from dereferencing freed glock objects, only free the
glock if the lockspace has actually been released.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Reviewed-by: Andrew Price <anprice@redhat.com>
GFS2 has been calling functions like dlm_lock() even after the lockspace
that these functions operate on has been released with
dlm_release_lockspace(). It has always assumed that those functions
would return -EINVAL in that case, but that was never guaranteed, and it
certainly is no longer the case since commit 4db41bf4f0 ("dlm: remove
ls_local_handle from struct dlm_ls").
To fix that, add proper lockspace locking.
Fixes: 3e11e53041 ("GFS2: ignore unlock failures after withdraw")
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Reviewed-by: Andrew Price <anprice@redhat.com>
Josua Mayer
Sarthak Garg
Tomáš Macholda
Monish Chunara
Ulf Hansson
Tomi Valkeinen
Krzysztof Kozlowski
AngeloGioacchino Del Regno
Geert Uytterhoeven
Duje Mihanović
Greg Kroah-Hartman
Guan-Yu Lin
Israel Cepeda
Linus Walleij
Thomas Fourier
Krishna Kurapati
Haotien Hsu
Kuen-Han Tsai
Herve Codina
Xu Yang
Edward Adam Davis
Thomas Huth
Ahelenia Ziemiańska
Georgi Djakov
Raviteja Laggyshetty
Zhang Enpei
Sumit Kumar
Michal Wajdeczko
Josh Poimboeuf
Andreas Gruenbacher