| Age | Commit message (Collapse) | Author |
|---|
|
Since commit 55b48e23f5c4 ("genirq/devres: Add error handling in
devm_request_*_irq()"), devm_request_irq() automatically logs
detailed error messages on failure. Remove the now-redundant
driver-specific dev_err() calls.
Signed-off-by: Pan Chuang <panchuang@vivo.com>
Reviewed-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Populating the BPMP firmware debugfs requires a sequence of synchronous
firmware transactions during BPMP probe, which impacts the overall boot
time. Currently, the only way to avoid populating the BPMP debugfs is to
disable CONFIG_DEBUG_FS completely.
Add a tegra_bpmp.debugfs module parameter so systems can skip populating
the BPMP debugfs with tegra_bpmp.debugfs=0. Keep it enabled by default to
preserve existing behavior. The parameter is defined in bpmp-debugfs.c,
so it is available only when CONFIG_DEBUG_FS is enabled.
Signed-off-by: Aniruddha Rao <anrao@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Different workloads can place different memory-bandwidth demands on the
system. Selecting an appropriate bandwidth limit depends on the runtime
workload mix and on the devices carrying that traffic, such as PCIe
traffic or GPU traffic connected over the chip-to-chip link (NVCLINK).
That information is not available to the kernel.
BPMP firmware of some Tegra platforms provides Memory Bandwidth
Throttler (MBWT) controls for PCIe and GPU traffic connected over
NVCLINK on the path to DRAM. Each PCIe bandwidth group has a single
shared cap for all traffic in that group. Bandwidth for a group can be
set per traffic type.
Describe the available bandwidth groups and traffic types in platform
data. Each entry maps a sysfs name to the corresponding firmware
instance and VC type so that these values can vary between platforms.
Add sysfs attributes on the tegra-bpmp platform device to expose a
narrow userspace interface for MBWT control. The attributes are arranged
as mbwt/<bandwidth-group>/<traffic-type>/bandwidth. The directory names
come from platform data. The current platform description uses pcie0
through pcie5 for bandwidth groups and pcie_read, pcie_write and
nvclink for traffic types.
Reading a bandwidth attribute queries firmware for the selected bandwidth
group and traffic type. Writing an integer programs the target bandwidth
cap in GB/s for that bandwidth group and traffic type.
Register the attributes only when BPMP firmware reports support for the
MBWT GET_BW and SET_BW requests through its query ABI.
Signed-off-by: Aniruddha Rao <anrao@nvidia.com>
[treding@nvidia.com: minor stylistic changes]
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add helper functions that send MBWT requests to BPMP firmware.
The helpers implement the GET_BW and SET_BW operations.
They also provide a QUERY_ABI-based probe for command availability.
Signed-off-by: Aniruddha Rao <anrao@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add required changes in the Tegra BPMP driver to make it compatible with
ACPI-based platforms.
On ACPI systems, IPC is handled through an AML method instead of the
kernel mailbox and IVC frameworks.
Clock, reset and power-domain providers are not controlled by the Linux
BPMP driver on ACPI-based systems, so do not initialize them there.
Keep transport initialization in the SoC-specific init operation. Use an
optional init_providers operation for DT child and provider registration
after BPMP communication has been established and the firmware tag has
been read. ACPI platforms leave this operation unset.
DT transports register an MRQ_PING handler for requests initiated by
BPMP firmware. This request direction was used only by a diagnostic
command in older firmware and is not provided by firmware on ACPI-based
systems. The ACPI transport therefore does not register an incoming
handler; its init operation only sends the outgoing MRQ_PING request used
to verify communication with firmware.
Signed-off-by: Aniruddha Rao <anrao@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Move BPMP channel allocation and initialization into a helper.
This keeps the probe function focused on sequencing.
It also lets later patches decide whether mailbox/IVC setup is needed.
No functional change is intended.
Signed-off-by: Aniruddha Rao <anrao@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
On multi-socket platforms each socket has its own BPMP that is
registered with the kernel, so the existing single fixed "bpmp"
debugfs directory name cannot accommodate more than one instance.
Group the per-socket BPMP debugfs entries under a shared top-level
/sys/kernel/debug/bpmp/ directory, with each socket's BPMP device
under a "<numa-node-id>-bpmp" subdirectory:
/sys/kernel/debug/bpmp/0-bpmp/...
/sys/kernel/debug/bpmp/1-bpmp/...
For a multi-socket platform, the root debugfs bpmp/ directory is created
by the first BPMP device that is populated. For single-socket platforms,
the existing directory structure is preserved.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The Tegra BPMP debugfs code returns -ENOMEM for most cases where calls
to debugfs_create_dir() or debugfs_create_file() fail. These debugfs
functions return an ERR_PTR with the actual error code on failure.
Therefore, update the Tegra BPMP debugfs code to propagate the actual
error code on failure.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
kconfiglint reports:
K002: config TEGRA_BPMP selects visible symbol TEGRA_IVC which has
dependencies
TEGRA_IVC was originally introduced in
commit ca791d7f4256 ("firmware: tegra:
Add IVC library") as a user-visible
bool with a prompt ("Tegra IVC protocol"). At that time, TEGRA_BPMP
depended on TEGRA_IVC, requiring users to manually enable it.
Recently,
commit 78eb18020a88 ("firmware: tegra: Fix IVC dependency problems")
recognized that TEGRA_IVC is library code that should be activated via
`select` rather than user selection. That commit changed TEGRA_BPMP from
`depends on TEGRA_IVC` to `select TEGRA_IVC`, and restricted TEGRA_IVC's
prompt to only appear under COMPILE_TEST
(`bool "Tegra IVC protocol" if COMPILE_TEST`). The commit message
explicitly states: "The IVC code is library code that other drivers need to
select if they need that library."
However, the `if COMPILE_TEST` qualifier still leaves TEGRA_IVC as a
technically visible symbol, triggering K002 when TEGRA_BPMP selects it.
Since TEGRA_IVC depends on ARCH_TEGRA, it cannot be independently enabled
under COMPILE_TEST without ARCH_TEGRA anyway, limiting the value of the
standalone COMPILE_TEST path. TEGRA_BPMP itself provides adequate
COMPILE_TEST coverage for the IVC library through its own dependency chain.
Complete the transition to a pure library symbol by removing the prompt
entirely, making TEGRA_IVC a hidden bool activated only via select from
TEGRA_BPMP. This is consistent with the intent expressed in 78eb18020a88.
Assisted-by: Claude:claude-opus-4-6 kconfiglint
Signed-off-by: Sasha Levin <sashal@kernel.org>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Some device tree bindings need to specify a parameter along with a BPMP
phandle reference to designate the ID associated with a given controller
that needs to interoperate with BPMP. Typically this is specified as an
extra cell in the nvidia,bpmp property, so add a helper to parse this ID
while resolving the phandle reference.
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The IPC shared memory can reside in system memory or SRAM. In the latter
case the memory-region property is expected not to be present, so do not
warn about it.
Reported-by: Jonathan Hunter <jonathanh@nvidia.com>
Fixes: dbe4efea38d0 ("firmware: tegra: bpmp: Use of_reserved_mem_region_to_resource() for "memory-region"")
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The definition of tegra186_bpmp_ops was not updated in sync with the use
in bpmp.c:
drivers/firmware/tegra/bpmp.c:856:17: error: 'tegra186_bpmp_ops' undeclared here (not in a function); did you mean 'tegra_bpmp_ops'?
856 | .ops = &tegra186_bpmp_ops,
aarch64-linux-ld: drivers/firmware/tegra/bpmp.o:(.rodata+0x2f0): undefined reference to `tegra186_bpmp_ops'
Update the Makefile as needed.
There is really no need to hide the declaration based on the configuration,
so just expose it unconditionally so it never has to be updated again
for the next SoC.
Fixes: 94bce2cf7cf6 ("firmware: tegra: bpmp: Add support on Tegra264")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Link: https://lore.kernel.org/r/20250711082409.1398497-1-arnd@kernel.org
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
"memory-region"
Use the newly added of_reserved_mem_region_to_resource() function to
handle "memory-region" properties.
Signed-off-by: Rob Herring (Arm) <robh@kernel.org>
Link: https://lore.kernel.org/r/20250703183434.2073375-1-robh@kernel.org
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Support for Tegra264 depends on the Tegra186 support, so make sure the
latter is enabled.
Link: https://lore.kernel.org/r/20250506133118.1011777-11-thierry.reding@gmail.com
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The IVC code is library code that other drivers need to select if they
need that library. However, if the symbol is user-selectable this can
lead to conflicts.
Fix this by making the symbol only selectable for COMPILE_TEST and add
a select TEGRA_IVC to TEGRA_BPMP, which is currently the only user.
Link: https://lore.kernel.org/r/20250506133118.1011777-10-thierry.reding@gmail.com
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
error paths"
This reverts commit 8812b8689ee6 ("firmware: tegra: bpmp: Use scoped
device node handling to simplify error paths") because it was silently
modified by committer during commit process, by moving declaration of
'struct device_node *np' above the initializer/constructor. Such code
was not intention of the author, is not conforming to cleanup.h code
style and decreases the code readability.
I did not write such code and I did not agree to put my name with such
commit.
Original patch:
https://lore.kernel.org/all/20240816135722.105945-2-krzysztof.kozlowski@linaro.org/
Cc: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
[treding@nvidia.com: shorten subject line]
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Obtain the device node reference with scoped/cleanup.h to reduce error
handling and make the code a bit simpler.
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
mbox_client_to_bpmp() is not used, W=1 builds:
drivers/firmware/tegra/bpmp.c:28:1: error: unused function 'mbox_client_to_bpmp' [-Werror,-Wunused-function]
Fixes: cdfa358b248e ("firmware: tegra: Refactor BPMP driver")
Cc: stable@vger.kernel.org
Signed-off-by: Krzysztof Kozlowski <krzysztof.kozlowski@linaro.org>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
get_filename()
The kfree() function was called in one case by
the get_filename() function during error handling
even if the passed variable contained a null pointer.
This issue was detected by using the Coccinelle software.
Thus return directly after a call of the function “kzalloc” failed
at the beginning.
Signed-off-by: Markus Elfring <elfring@users.sourceforge.net>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc
Pull SoC driver updates from Arnd Bergmann:
"The highlights for the driver support this time are
- Qualcomm platforms gain support for the Qualcomm Secure Execution
Environment firmware interface to access EFI variables on certain
devices, and new features for multiple platform and firmware
drivers.
- Arm FF-A firmware support gains support for v1.1 specification
features, in particular notification and memory transaction
descriptor changes.
- SCMI firmware support now support v3.2 features for clock and DVFS
configuration and a new transport for Qualcomm platforms.
- Minor cleanups and bugfixes are added to pretty much all the active
platforms: qualcomm, broadcom, dove, ti-k3, rockchip, sifive,
amlogic, atmel, tegra, aspeed, vexpress, mediatek, samsung and
more.
In particular, this contains portions of the treewide conversion to
use __counted_by annotations and the device_get_match_data helper"
* tag 'soc-drivers-6.7' of git://git.kernel.org/pub/scm/linux/kernel/git/soc/soc: (156 commits)
soc: qcom: pmic_glink_altmode: Print return value on error
firmware: qcom: scm: remove unneeded 'extern' specifiers
firmware: qcom: scm: add a missing forward declaration for struct device
firmware: qcom: move Qualcomm code into its own directory
soc: samsung: exynos-chipid: Convert to platform remove callback returning void
soc: qcom: apr: Add __counted_by for struct apr_rx_buf and use struct_size()
soc: qcom: pmic_glink: fix connector type to be DisplayPort
soc: ti: k3-socinfo: Avoid overriding return value
soc: ti: k3-socinfo: Fix typo in bitfield documentation
soc: ti: knav_qmss_queue: Use device_get_match_data()
firmware: ti_sci: Use device_get_match_data()
firmware: qcom: qseecom: add missing include guards
soc/pxa: ssp: Convert to platform remove callback returning void
soc/mediatek: mtk-mmsys: Convert to platform remove callback returning void
soc/mediatek: mtk-devapc: Convert to platform remove callback returning void
soc/loongson: loongson2_guts: Convert to platform remove callback returning void
soc/litex: litex_soc_ctrl: Convert to platform remove callback returning void
soc/ixp4xx: ixp4xx-qmgr: Convert to platform remove callback returning void
soc/ixp4xx: ixp4xx-npe: Convert to platform remove callback returning void
soc/hisilicon: kunpeng_hccs: Convert to platform remove callback returning void
...
|
|
Add suspend hook and a 'suspended' field in the 'struct tegra_bpmp'
to mark if BPMP is suspended. Also, add a 'flags' field in the
'struct tegra_bpmp_message' whose 'TEGRA_BPMP_MESSAGE_RESET' bit can be
set from the Tegra MC driver to signal that the reset of BPMP IPC
channels is required before sending MRQ to the BPMP FW. Together both
the fields allow us to handle any requests that might be sent too soon
as they can cause hang during system resume.
One case where we see BPMP requests being sent before the BPMP driver
has resumed is the memory bandwidth requests which are triggered by
onlining the CPUs during system resume. The CPUs are onlined before the
BPMP has resumed and we need to reset the BPMP IPC channels to handle
these requests.
The additional check for 'flags' is done to avoid any un-intended BPMP
IPC reset if the tegra_bpmp_transfer*() API gets called during suspend
sequence after the BPMP driver is suspended.
Fixes: f41e1442ac5b ("cpufreq: tegra194: add OPP support and set bandwidth")
Co-developed-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Sumit Gupta <sumitg@nvidia.com>
Acked-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
`strncpy` is deprecated for use on NUL-terminated destination strings [1].
We should prefer more robust and less ambiguous string interfaces.
It seems like the filename stored at `namevirt` is expected to be
NUL-terminated.
A suitable replacement is `strscpy_pad` due to the fact that it
guarantees NUL-termination on the destination buffer whilst maintaining
the NUL-padding behavior that strncpy provides.
Link: https://www.kernel.org/doc/html/latest/process/deprecated.html#strncpy-on-nul-terminated-strings [1]
Link: https://github.com/KSPP/linux/issues/90
Cc: linux-hardening@vger.kernel.org
Signed-off-by: Justin Stitt <justinstitt@google.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Link: https://lore.kernel.org/r/20230913-strncpy-drivers-firmware-tegra-bpmp-debugfs-c-v1-1-828b0a8914b5@google.com
Signed-off-by: Kees Cook <keescook@chromium.org>
|
|
The DT of_device.h and of_platform.h date back to the separate
of_platform_bus_type before it as merged into the regular platform bus.
As part of that merge prepping Arm DT support 13 years ago, they
"temporarily" include each other. They also include platform_device.h
and of.h. As a result, there's a pretty much random mix of those include
files used throughout the tree. In order to detangle these headers and
replace the implicit includes with struct declarations, users need to
explicitly include the correct includes.
Acked-by: Dinh Nguyen <dinguyen@kernel.org>
Acked-by: Thierry Reding <treding@nvidia.com>
Signed-off-by: Rob Herring <robh@kernel.org>
Acked-by: Sudeep Holla <sudeep.holla@arm.com>
Link: https://lore.kernel.org/r/20230803-dt-header-cleanups-for-soc-v2-17-d8de2cc88bff@kernel.org
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
Implement support for DRAM MRQ GSCs.
Signed-off-by: Peter De Schrijver <pdeschrijver@nvidia.com>
[treding@nvidia.com: drop unnecessary discrimination enum]
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
It is preferred to use typed property access functions (i.e.
of_property_read_<type> functions) rather than low-level
of_get_property/of_find_property functions for reading properties. As
part of this, convert of_get_property/of_find_property calls to the
recently added of_property_present() helper when we just want to test
for presence of a property and nothing more.
Signed-off-by: Rob Herring <robh@kernel.org>
Acked-by: Sudeep Holla <sudeep.holla@arm.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Some error paths in mrq_debug_read and bpmp_debug_show would overwrite
the return error code with a subsequent call to mrq_debug_close.
Change the code to only change the error code if there was no prior
error.
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The BPMP firmware driver never worked on big-endian kernels, and
cannot easily be made portable. Add a dependency to make this clear
in case anyone ever wants to try a big-endian kernel on this hardware.
Link: https://lore.kernel.org/linux-arm-kernel/Y34FCQ3xTmcjqKRT@orome/
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
|
|
There is no need to call the dev_err() function directly to print a
custom message when handling an error from either the platform_get_irq()
or platform_get_irq_byname() functions as both are going to display an
appropriate error message in case of a failure.
./drivers/firmware/tegra/bpmp-tegra210.c:204:2-9: line 204 is redundant
because platform_get_irq() already prints an error
./drivers/firmware/tegra/bpmp-tegra210.c:216:2-9: line 216 is redundant
because platform_get_irq() already prints an error
Link: https://bugzilla.openanolis.cn/show_bug.cgi?id=2579
Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Signed-off-by: Yang Li <yang.lee@linux.alibaba.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The CPU and BPMP inter-processor communication code is only partially
endian-aware, so it doesn't work properly when run on a big-endian CPU
anyway. Running Tegra SoCs in big-endian mode has also never been
supported, especially not on those with 64-bit ARM processors.
If big-endian support ever becomes necessary this can be added back but
will need additional fixes for completeness.
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The shared memory used for inter-processor communication between the CPU
and the BPMP can reside either in system memory or in I/O memory. Use
the iosys-map helpers to abstract these differences away.
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The canonical type for 32-bit unsigned integers in the kernel is u32, so
use that instead of uint32_t.
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
This patch fixes the following Coccinelle warning:
drivers/firmware/tegra/bpmp-debugfs.c:379: WARNING opportunity for memdup_user
Use memdup_user() rather than duplicating its implementation. This is a
little bit restricted to reduce false positives.
Signed-off-by: Qing Wang <wangqing@vivo.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
If an error occurs, debugfs_create_file() will return ERR_PTR(-ERROR),
so use IS_ERR() to check it.
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Lv Ruyi <lv.ruyi@zte.com.cn>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Use memcpy_toio and memcpy_fromio variants of memcpy to guarantee no
unaligned access to IPC memory area. This is to allow the IPC memory to
be mapped as Device memory to further suppress speculative reads from
happening within the 64 kB memory area above the IPC memory when 64 kB
memory pages are used.
Signed-off-by: Timo Alho <talho@nvidia.com>
Signed-off-by: Mikko Perttunen <mperttunen@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Application of sizeof() to pointer yields the number of bytes of the
pointer, but it should use the length of buffer in the code.
Fixes: 06c2d9a078ab ("firmware: tegra: Reduce stack usage")
Reported-by: Zeal Robot <zealci@zte.com.cn>
Signed-off-by: Lv Ruyi <lv.ruyi@zte.com.cn>
Reviewed-by: Jon Hunter <jonathanh@nvidia.com>
Tested-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Use the devm_platform_ioremap_resource() helper instead of calling
platform_get_resource() and devm_ioremap_resource() separately.
Signed-off-by: Cai Huoqing <caihuoqing@baidu.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Building the bpmp-debugfs driver for Arm results in a warning for stack usage:
drivers/firmware/tegra/bpmp-debugfs.c:321:16: error: stack frame size of 1224 bytes in function 'bpmp_debug_store' [-Werror,-Wframe-larger-than=]
static ssize_t bpmp_debug_store(struct file *file, const char __user *buf,
It should be possible to rearrange the code to not require two separate
buffers for the file name, but the easiest workaround is to use dynamic
allocation.
Fixes: 5e37b9c137ee ("firmware: tegra: Add support for in-band debug")
Link: https://lore.kernel.org/all/20201204193714.3134651-1-arnd@kernel.org/
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
[treding@nvidia.com: consistently return NULL on failure]
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Opencode a copy of mrq_debug_read() in bpmp_debug_show() so that it
can use seq_write() directly instead of poking holes into the seq_file
abstractions using seq_get_buf().
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The tegra186_bpmp_ops symbol is used on Tegra234, so make sure it's
available.
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
When call irq_get_irq_data() to get the IRQ's irq_data failed, an
appropriate error code -ENOENT should be returned. However, we directly
return 'err', which records the IRQ number instead of the error code.
Fixes: 139251fc2208 ("firmware: tegra: add bpmp driver for Tegra210")
Reported-by: Hulk Robot <hulkci@huawei.com>
Signed-off-by: Zhen Lei <thunder.leizhen@huawei.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The way that bpmp_populate_debugfs_inband() uses strncpy()
and strncat() makes no sense since the size argument for
the first is insufficient to contain the trailing '/'
and the second passes the length of the input rather than
the output, which triggers a warning:
In function 'strncat',
inlined from 'bpmp_populate_debugfs_inband' at ../drivers/firmware/tegra/bpmp-debugfs.c:422:4:
include/linux/string.h:289:30: warning: '__builtin_strncat' specified bound depends on the length of the source argument [-Wstringop-overflow=]
289 | #define __underlying_strncat __builtin_strncat
| ^
include/linux/string.h:367:10: note: in expansion of macro '__underlying_strncat'
367 | return __underlying_strncat(p, q, count);
| ^~~~~~~~~~~~~~~~~~~~
drivers/firmware/tegra/bpmp-debugfs.c: In function 'bpmp_populate_debugfs_inband':
include/linux/string.h:288:29: note: length computed here
288 | #define __underlying_strlen __builtin_strlen
| ^
include/linux/string.h:321:10: note: in expansion of macro '__underlying_strlen'
321 | return __underlying_strlen(p);
Simplify this to use an snprintf() instead.
Fixes: 5e37b9c137ee ("firmware: tegra: Add support for in-band debug")
Signed-off-by: Arnd Bergmann <arnd@arndb.de>
Acked-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Enable support for the BPMP on Tegra234 to avoid relying on Tegra194
being enabled to pull in the needed OF device ID table entry.
On simulation platforms the BPMP hasn't booted up yet by the time we
probe the BPMP driver and the BPMP hasn't had a chance to mark the
doorbell as ringable by the CCPLEX. This corresponding check in the
BPMP driver will therefore fail. Work around this by disabling the
check on simulation platforms.
Reviewed-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add support for retrieving BPMP debug information via in-band messaging
as opposed to using shared-memory which older BPMP firmware used. Note
that it is possible to detect at runtime whether the BPMP firmware being
used supports the in-band messaging for retrieving the debug
informaation. Therefore, if the BPMP firmware supports the in-band
messaging for debug use this and otherwise fall-back to using shared
memory.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Currently, BPMP debug information is accessible via the Linux debugfs
file-system using a shared-memory scheme. More recent BPMP firmware now
supports accessing the debug information by in-band messaging which does
not require shared-memory. To prepare for adding in-band debugfs support
for the BPMP, move the shared-memory specific initialisation from the
tegra_bpmp_init_debugfs() into a sub-function.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Most functions in the BPMP driver use 'err' as the return variable
name but there are a few places that use 'ret'. Let's use 'err' to
be consistent.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Add checking of the BPMP-FW return code values for MRQ_DEBUGFS calls.
Also, development versions of the firmware may have debugfs with a
directory structure larger than 256 KiB. Hence increase the size of the
memory buffer to accommodate those firmware revisions.
And finally, ensure that no access outside of allocated memory buffer
happens in case BPMP-FW returns an invalid response size (nbytes) from
mrq_debugfs_dumpdir() call.
Signed-off-by: Timo Alho <talho@nvidia.com>
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Since commit 93d2e4322aa7 ("of: platform: Batch fwnode parsing when
adding all top level devices") was added, the probing of the Tegra
SRAM device has occurred later in the boot sequence, after the BPMP
has been probed. The BPMP uses sections of the SRAM for shared memory
and if the BPMP is probed before the SRAM then it fails to probe and
never tries again. This is causing a boot failure on Tegra186 and
Tegra194. Fix this by allowing the probe of the BPMP to be deferred if
the SRAM is not available yet.
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
The Tegra BPMP driver typically ends up deferring probe because it wants
to attach to the SMMU, so there's little sense in registering it at the
core init-level.
One side-effect of this is that the driver will be probed later even if
it doesn't want to attach to an SMMU, which means that consumers will
end up deferring probe, which in turn takes care of ordering the suspend
and resume queue in the correct way. Currently since suspend/resume
order depends on instantiation order, and because BPMP is listed at the
very end of the device tree (after most of its consumers), the suspend
and resume queue is ordered wrongly, which can cause issues for drivers
(like I2C) which suspend after and resume before BPMP. In the case of
I2C this typically leads to the clock failing to enable.
Besides fixing this suspend/resume ordering issue, this also has the
added benefit of allowing the driver to be built as a loadable module,
which can help decrease the size of multiplatform kernel.
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
A 'n' is mising in 'commuication'
Signed-off-by: Christophe JAILLET <christophe.jaillet@wanadoo.fr>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
|
Modules like PCIe in Tegra194 need BPMP firmware services in noirq phase
and hence move BPMP resume to noirq phase.
This patch is verified on Tegra210, Tegra186 and Tegra194.
Signed-off-by: Vidya Sagar <vidyas@nvidia.com>
Reviewed-by: Timo Alho <talho@nvidia.com>
Signed-off-by: Thierry Reding <treding@nvidia.com>
|
