Suzuki writes:
coresight: Updates for v6.5
CoreSight and hwtracing subsystem updates for v6.5 includes:
- Fixes to the CTI module reference leaks. This involves,
redesign of how the helper devices are tracked and CTI
devices have been converted to helper devices.
- Fix removal of the trctraceidr file from sysfs for ETMs.
- Match all ETMv4 instances based on the ETMv4 architected
registers and the CoreSight Component ID (CID), than having
to add individual PIDs for CPUs.
- Add support for Dummy CoreSight source and sink drivers.
- Add James Clark as Reviewer for the CoreSight kernel drivers
- Fixes to HiSilicon PCIe Tune and Trace Device driver
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
* tag 'coresight-next-v6.5' of git://git.kernel.org/pub/scm/linux/kernel/git/coresight/linux: (27 commits)
hwtracing: hisi_ptt: Fix potential sleep in atomic context
hwtracing: hisi_ptt: Advertise PERF_PMU_CAP_NO_EXCLUDE for PTT PMU
hwtracing: hisi_ptt: Export available filters through sysfs
hwtracing: hisi_ptt: Add support for dynamically updating the filter list
hwtracing: hisi_ptt: Factor out filter allocation and release operation
coresight: dummy: Update type of mode parameter in dummy_{sink,source}_enable()
Documentation: trace: Add documentation for Coresight Dummy Trace
dt-bindings: arm: Add support for Coresight dummy trace
Coresight: Add coresight dummy driver
MAINTAINERS: coresight: Add James Clark as Reviewer
coresight: etm4x: Match all ETM4 instances based on DEVARCH and DEVTYPE
coresight: etm4x: Make etm4_remove_dev() return void
coresight: etm4x: Fix missing trctraceidr file in sysfs
coresight: Fix CTI module refcount leak by making it a helper device
coresight: Enable and disable helper devices adjacent to the path
coresight: Refactor out buffer allocation function for ETR
coresight: Make refcount a property of the connection
coresight: Store in-connections as well as out-connections
coresight: Simplify connection fixup mechanism
coresight: Store pointers to connections rather than an array of them
...
Georgi writes:
interconnect changes for 6.5
This pull request contains the interconnect changes for the 6.5-rc1 merge
window which is a mix of core and driver changes with the following highlights:
- Support for configuring QoS on the Qualcomm's RPM-based platforms, that
required special handling of some interface (non-scaling) clocks.
- Support for clock-based interconnect providers for cases when clock
corresponds to bus bandwidth. This is used to enable CPU cluster bandwidth
scaling on MSM8996 platforms. One patch is touching a file in the clock
subsystem that has been acked by the maintainer.
Core changes:
interconnect: add clk-based icc provider support
interconnect: icc-clk: fix modular build
interconnect: drop unused icc_get() interface
Driver changes:
interconnect: qcom: rpm: Rename icc desc clocks to bus_blocks
interconnect: qcom: rpm: Rename icc provider num_clocks to num_bus_clocks
interconnect: qcom: rpm: Drop unused parameters
interconnect: qcom: rpm: Set QoS registers only once
interconnect: qcom: rpm: Handle interface clocks
interconnect: qcom: icc-rpm: Enforce 2 or 0 bus clocks
interconnect: qcom: rpm: Don't use clk_get_optional for bus clocks anymore
interconnect: qcom: msm8996: Promote to core_initcall
interconnect: qcom: rpm: allocate enough data in probe()
dt-bindings: interconnect/msm8996-cbf: add defines to be used by CBF
clk: qcom: cbf-msm8996: scale CBF clock according to the CPUfreq
dt-bindings: interconnect: fsl,imx8m-noc: drop unneeded quotes
Signed-off-by: Georgi Djakov <djakov@kernel.org>
* tag 'icc-6.5-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/djakov/icc:
dt-bindings: interconnect: fsl,imx8m-noc: drop unneeded quotes
interconnect: icc-clk: fix modular build
clk: qcom: cbf-msm8996: scale CBF clock according to the CPUfreq
interconnect: drop unused icc_get() interface
interconnect: qcom: rpm: allocate enough data in probe()
interconnect: qcom: msm8996: Promote to core_initcall
interconnect: qcom: rpm: Don't use clk_get_optional for bus clocks anymore
interconnect: qcom: icc-rpm: Enforce 2 or 0 bus clocks
interconnect: qcom: rpm: Handle interface clocks
interconnect: add clk-based icc provider support
dt-bindings: interconnect/msm8996-cbf: add defines to be used by CBF
interconnect: qcom: rpm: Set QoS registers only once
interconnect: qcom: rpm: Drop unused parameters
interconnect: qcom: rpm: Rename icc provider num_clocks to num_bus_clocks
interconnect: qcom: rpm: Rename icc desc clocks to bus_blocks
We're using pci_irq_vector() to obtain the interrupt number and then
bind it to the CPU start perf under the protection of spinlock in
pmu::start(). pci_irq_vector() might sleep since [1] because it will
call msi_domain_get_virq() to get the MSI interrupt number and it
needs to acquire dev->msi.data->mutex. Getting a mutex will sleep on
contention. So use pci_irq_vector() in an atomic context is problematic.
This patch cached the interrupt number in the probe() and uses the
cached data instead to avoid potential sleep.
[1] commit 82ff8e6b78 ("PCI/MSI: Use msi_get_virq() in pci_get_vector()")
Fixes: ff0de066b4 ("hwtracing: hisi_ptt: Add trace function support for HiSilicon PCIe Tune and Trace device")
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20230621092804.15120-6-yangyicong@huawei.com
The PTT trace collects PCIe TLP headers from the PCIe link and don't
have the ability to exclude certain context. It doesn't support itrace
as well. So replace PERF_PMU_CAP_ITRACE with PERF_PMU_CAP_NO_EXCLUDE.
This will greatly save the storage of final data. Tested tracing idle
link for ~15s, without this patch we'll collect ~28.682MB data for
additional information and with this patch it reduced to ~0.226MB.
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Tested-by: Junhao He <hejunhao3@huawei.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20230621092804.15120-5-yangyicong@huawei.com
The PTT can only filter the traced TLP headers by the Root Ports or the
Requester ID of the Endpoint, which are located on the same PCIe core of
the PTT device. The filter value used is derived from the BDF number of
the supported Root Port or the Endpoint. It's not friendly enough for the
users since it requires the user to be familiar enough with the platform
and calculate the filter value manually.
This patch export the available filters through sysfs. Each available
filters is presented as an individual file with the name of the BDF
number of the related PCIe device. The files are created under
$(PTT PMU dir)/available_root_port_filters and
$(PTT PMU dir)/available_requester_filters respectively. The filter
value can be known by reading the related file.
Then the users can easily know the available filters for trace and get
the filter values without calculating.
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20230621092804.15120-4-yangyicong@huawei.com
The PCIe devices supported by the PTT trace can be removed/rescanned by
hotplug or through sysfs. Add support for dynamically updating the
available filter list by registering a PCI bus notifier block. Then user
can always get latest information about available tracing filters and
driver can block the invalid filters of which related devices no longer
exist in the system.
Reviewed-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20230621092804.15120-3-yangyicong@huawei.com
Chanwoo writes:
Update extcon next for v6.5
Detailed description for this pull request:
1. Clean-up extcon core without any behavior changes
- Add extcon_alloc_cables/muex/groups to improve the readability
of extcon_dev_register.
- Fix kernel doc of property and property capability fields to aovid warnings
and add missing description of struct extcon_dev.
- Use DECLARE_BITMAP macro and sysfs_emit instead of sprintf
- Use device_match_of_node helper instead of accessing the .of_node
- Use ida_alloc/free to get the unique id for extcon device
2. Update extcon-usbc-tusb320.c to support usb_role_switch and accessory detection
- Add usb_role_switch support on extcon-usbsc-tusb320.
- Add additional accessory detection for audio/debug accessory
and then pass the deteced accessory information to typec subsystem
on extcon-usbsc-tusb320.c.
- Add the support of unregistration of typec port on both error handling
and driver removal step on
3. Update extcon provider drivers (apx288/qcom-spmi-misc/palmas)
- Replace put_device with acpi_dev_put on extcon-axp288.c
- Use platform_get_irq_byname_optional for getting irq of
usb_id and usb_vbus on extcon-qcom-spmi-misc.c.
- Remove unused of_gpio.h on extcon-palmas.c.
4. Fix the devicetree binding document
- Rename misc node name to 'usb-dect@900' on pm8941-misc.yam
- Fix usb-id and usb_vbus defintion on pm8941-misc.yaml
- Drop unneeded quotes from extcon-arizona.c devicetree documentation
* tag 'extcon-next-for-6.5' of git://git.kernel.org/pub/scm/linux/kernel/git/chanwoo/extcon: (26 commits)
dt-bindings: extcon: wlf,arizona: drop unneeded quotes
extcon: Switch i2c drivers back to use .probe()
extcon: Drop unneeded assignments
extcon: Use sizeof(*pointer) instead of sizeof(type)
extcon: Use unique number for the extcon device ID
extcon: Remove dup device name in the message and unneeded error check
extcon: Use dev_of_node(dev) instead of dev->of_node
extcon: Use device_match_of_node() helper
extcon: Amend kernel documentation of struct extcon_dev
extcon: Use sysfs_emit() to instead of sprintf()
extcon: Use DECLARE_BITMAP() to declare bit arrays
extcon: Fix kernel doc of property capability fields to avoid warnings
extcon: Fix kernel doc of property fields to avoid warnings
extcon: usbc-tusb320: add USB_ROLE_SWITCH dependency
extcon: usbc-tusb320: add usb_role_switch support
extcon: usbc-tusb320: add accessory detection support
extcon: Add extcon_alloc_groups to simplify extcon register function
extcon: Add extcon_alloc_muex to simplify extcon register function
extcon: Add extcon_alloc_cables to simplify extcon register function
extcon: Remove redundant null checking for class
...
Clang's kernel Control Flow Integrity (kCFI) is a compiler-based
security mitigation that ensures the target of an indirect function call
matches the expected type of the call and trapping if they do not match
exactly. The warning -Wincompatible-function-pointer-types-strict aims
to catch these issues at compile time, which reveals:
drivers/hwtracing/coresight/coresight-dummy.c:53:12: error: incompatible function pointer types initializing 'int (*)(struct coresight_device *, struct perf_event *, enum cs_mode)' with an expression of type 'int (struct coresight_device *, struct perf_event *, u32)' (aka 'int (struct coresight_device *, struct perf_event *, unsigned int)') [-Werror,-Wincompatible-function-pointer-types-strict]
53 | .enable = dummy_source_enable,
| ^~~~~~~~~~~~~~~~~~~
drivers/hwtracing/coresight/coresight-dummy.c:62:12: error: incompatible function pointer types initializing 'int (*)(struct coresight_device *, enum cs_mode, void *)' with an expression of type 'int (struct coresight_device *, u32, void *)' (aka 'int (struct coresight_device *, unsigned int, void *)') [-Werror,-Wincompatible-function-pointer-types-strict]
62 | .enable = dummy_sink_enable,
| ^~~~~~~~~~~~~~~~~
2 errors generated.
Commit 9fa3682869 ("coresight: Use enum type for cs_mode wherever
possible") updated the type of the mode parameter in the prototype but
this driver was not introduced until commit 9d3ba0b6c0 ("Coresight:
Add coresight dummy driver") and 'int' is ABI compatible with 'enum
cs_mode', so there is no warning from regular
-Wincompatible-function-pointer-types.
Adjust the type of the mode parameter in the callback implementations in
the coresight dummy driver to match the prototype, clearing up the
warning and avoiding kCFI failures at runtime.
Fixes: 9d3ba0b6c0 ("Coresight: Add coresight dummy driver")
Signed-off-by: Nathan Chancellor <nathan@kernel.org>
Signed-off-by: Suzuki K Poulose <suzuki.poulose@arm.com>
Link: https://lore.kernel.org/r/20230616-coresight-dummy-fix-kcfi-warnings-v1-1-c55c64f8f0f5@kernel.org
The .remove() callback for a platform driver returns an int which makes
many driver authors wrongly assume it's possible to do error handling by
returning an error code. However the value returned is ignored (apart
from emitting a warning) and this typically results in resource leaks.
To improve here there is a quest to make the remove callback return
void. In the first step of this quest all drivers are converted to
.remove_new() which already returns void. Eventually after all drivers
are converted, .remove_new() is renamed to .remove().
Trivially convert this driver from always returning zero in the remove
callback to the void returning variant.
Signed-off-by: Uwe Kleine-König <u.kleine-koenig@pengutronix.de>
Acked-by: Michal Simek <michal.simek@amd.com>
Message-ID: <20230605092047.50472-3-u.kleine-koenig@pengutronix.de>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
A process can spawn a PD on DSP with some attributes that can be
associated with the PD during spawn and run. The invocation
corresponding to the create request with attributes has total
4 buffers at the DSP side implementation. If this number is not
correct, the invocation is expected to fail on DSP. Added change
to use correct number of buffer count for creating fastrpc scalar.
Fixes: d73f71c7c6 ("misc: fastrpc: Add support for create remote init process")
Cc: stable <stable@kernel.org>
Tested-by: Ekansh Gupta <quic_ekangupt@quicinc.com>
Signed-off-by: Ekansh Gupta <quic_ekangupt@quicinc.com>
Message-ID: <1686743685-21715-1-git-send-email-quic_ekangupt@quicinc.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In a couple of situations like
name = kstrndup(buf, count, GFP_KERNEL);
if (!name)
return -ENOSPC;
the error is not actually "No space left on device", but "Out of memory".
It is semantically correct to return -ENOMEM in all failed kstrndup()
and kzalloc() cases in this driver, as it is not a problem with disk
space, but with kernel memory allocator failing allocation.
The semantically correct should be:
name = kstrndup(buf, count, GFP_KERNEL);
if (!name)
return -ENOMEM;
Cc: Dan Carpenter <error27@gmail.com>
Cc: Takashi Iwai <tiwai@suse.de>
Cc: Kees Cook <keescook@chromium.org>
Cc: "Luis R. Rodriguez" <mcgrof@ruslug.rutgers.edu>
Cc: Scott Branden <sbranden@broadcom.com>
Cc: Hans de Goede <hdegoede@redhat.com>
Cc: Brian Norris <briannorris@chromium.org>
Fixes: c92316bf8e ("test_firmware: add batched firmware tests")
Fixes: 0a8adf5847 ("test: add firmware_class loader test")
Fixes: 548193cba2 ("test_firmware: add support for firmware_request_platform")
Fixes: eb910947c8 ("test: firmware_class: add asynchronous request trigger")
Fixes: 061132d2b9 ("test_firmware: add test custom fallback trigger")
Fixes: 7feebfa487 ("test_firmware: add support for request_firmware_into_buf")
Signed-off-by: Mirsad Goran Todorovac <mirsad.todorovac@alu.unizg.hr>
Reviewed-by: Dan Carpenter <dan.carpenter@linaro.org>
Message-ID: <20230606070808.9300-1-mirsad.todorovac@alu.unizg.hr>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
This adds support for the "fixed-layout" NVMEM layout binding. It allows
defining NVMEM cells in a layout DT node named "nvmem-layout".
While NVMEM subsystem supports layout drivers it has been discussed that
"fixed-layout" may actually be supperted internally. It's because:
1. It's a very basic layout
2. It allows sharing code with legacy syntax parsing
3. It's safer for soc_device_match() due to -EPROBE_DEFER
4. This will make the syntax transition easier
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Reviewed-by: Michael Walle <michael@walle.cc>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Message-ID: <20230611140330.154222-26-srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
With the introduction of NVMEM layouts, new NVMEM content structures
should be defined as such. We should also try to convert / migrate
existing NVMEM content bindings to layouts.
This commit handles fixed NVMEM cells. So far they had to be defined
directly - as device subnodes. With this change it's allowed to put them
in the DT node named "nvmem-layout".
Having NVMEM cells in separated node is preferred as it draws a nice
line between NVMEM device and its content. It results in cleaner
bindings.
FWIW a very similar situation has happened to MTD devices and their
partitions: see commit 5d96ea42eb ("dt-bindings: mtd: Clarify all
partition subnodes").
Signed-off-by: Rafał Miłecki <rafal@milecki.pl>
Reviewed-by: Rob Herring <robh@kernel.org>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Message-ID: <20230611140330.154222-24-srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Add i.MX93 OCOTP support. i.MX93 OCOTP has two parts: Fuse shadow
block(fsb) and fuse managed by ELE. The FSB part could be directly
accessed with MMIO, the ELE could only be accessed with ELE API.
Currently the ELE API is not ready, so NULL function callback is used,
but it was tested with downstream ELE API.
Signed-off-by: Peng Fan <peng.fan@nxp.com>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Message-ID: <20230611140330.154222-22-srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
In preparation to support new Rockchip OTP memory devices with different
clock configurations and register layout, extend rockchip_data struct
with the related members: clks, num_clks, reg_read.
Additionally, to avoid managing redundant driver data, drop num_clks
member from rockchip_otp struct and update all references to point to
the equivalent member in rockchip_data.
Signed-off-by: Cristian Ciocaltea <cristian.ciocaltea@collabora.com>
Tested-by: Vincent Legoll <vincent.legoll@gmail.com>
Reviewed-by: Heiko Stuebner <heiko@sntech.de>
Signed-off-by: Srinivas Kandagatla <srinivas.kandagatla@linaro.org>
Message-ID: <20230611140330.154222-10-srinivas.kandagatla@linaro.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
