Kartik Rajput [Tue, 24 Mar 2026 05:58:42 +0000 (11:28 +0530)]
i2c: tegra: Add logic to support different register offsets
Tegra410 use different offsets for existing I2C registers, update
the logic to use appropriate offsets per SoC.
As the register offsets are now defined in the SoC-specific
tegra_i2c_regs structures, the tegra_i2c_reg_addr() function is no
longer needed to translate register offsets and has been removed.
Kartik Rajput [Tue, 24 Mar 2026 05:58:41 +0000 (11:28 +0530)]
i2c: tegra: Introduce tegra_i2c_variant to identify DVC and VI
Replace the per-instance DVC/VI boolean flags with a tegra_i2c_variant
enum and move the variant field into tegra_i2c_hw_feature so it is
populated via SoC match data.
Add dedicated SoC data entries for the "nvidia,tegra20-i2c-dvc" and
"nvidia,tegra210-i2c-vi" compatibles and drop compatible-string checks
from tegra_i2c_parse_dt.
Suggested-by: Jon Hunter <jonathanh@nvidia.com> Signed-off-by: Kartik Rajput <kkartik@nvidia.com> Reviewed-by: Jon Hunter <jonathanh@nvidia.com> Tested-by: Jon Hunter <jonathanh@nvidia.com> Signed-off-by: Andi Shyti <andi.shyti@kernel.org> Link: https://lore.kernel.org/r/20260324055843.549808-2-kkartik@nvidia.com
Stephen Brennan [Fri, 20 Mar 2026 23:45:53 +0000 (16:45 -0700)]
perf report: Add comm_nodigit sort key
The "comm" column allows grouping events by the process command. It is
intended to group like programs, despite having different PIDs. But some
workloads may adjust their own command, so that a unique identifier
(e.g. a PID or some other numeric value) is part of the command name.
This destroys the utility of "comm", forcing perf to place each unique
process name into its own bucket, which can contribute to a
combinatorial explosion of memory use in perf report.
Create a less strict version of this column, which ignores digits when
comparing command names. Commands whose names are the same (ignoring
digits) are sorted into the same histogram buckets, and displayed with
the placeholder value "<N>" in the place of digits. For example,
hypothetical command names "kworker/1" "kworker/2" "kworker/3" would
sort into the same bucket and be represented as "kworker/<N>".
Ian Rogers [Thu, 19 Mar 2026 23:33:49 +0000 (16:33 -0700)]
perf stat: Fix opt->value type for parse_cache_level
Commit f5803651b4a4 ("perf stat: Choose the most disaggregate command
line option") changed aggregation option handling for `perf stat` but
not `perf stat report` leading to parse_cache_level being passed a
struct in the `perf stat` case but erroneously an aggr_mode enum value
for `perf stat report`. Change the `perf stat report` aggregation
handling to use the same opt_aggr_mode as `perf stat`. Also, just pass
the boolean for consistency with other boolean argument handling.
Fixes: f5803651b4a4 ("perf stat: Choose the most disaggregate command line option") Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Ian Rogers [Thu, 19 Mar 2026 23:33:48 +0000 (16:33 -0700)]
perf lock: Fix option value type in parse_max_stack
The value is a void* and the address of an int, max_stack_depth, is
set up in the perf lock options. The parse_max_stack function treats
the int* as a long*, make this more correct by declaring the value to
be an int*.
Fixes: 0a277b622670 ("perf lock contention: Check --max-stack option") Signed-off-by: Ian Rogers <irogers@google.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
Thomas Richter [Thu, 19 Mar 2026 08:47:54 +0000 (09:47 +0100)]
perf record: Add support for arch_sdt_arg_parse_op() on s390
commit e5e66adfe45a6 ("perf regs: Remove __weak attributive arch_sdt_arg_parse_op() function")
removes arch_sdt_arg_parse_op() functions and reveals missing s390 support.
The following warning is printed:
Unknown ELF machine 22, standard arguments parse will be skipped.
ELF machine 22 is the EM_S390 host. This happens with command
# ./perf record -v -- stress-ng -t 1s --matrix 0
when the event is not specified.
Add s390 specific __perf_sdt_arg_parse_op_s390() function to support
-architecture calls to arch_sdt_arg_parse_op() for s390.
The warning disappears.
Signed-off-by: Thomas Richter <tmricht@linux.ibm.com> Reviewed-by: Ian Rogers <irogers@google.com> Tested-by: Jan Polensky <japo@linux.ibm.com> Cc: Dapeng Mi <dapeng1.mi@linux.intel.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
The USB ID 2040:8360 was reported as a new revision
of Hauppauge HVR-935 and requested to be added to
the cx231xx driver. The issue is current generation
of Hauppauge 9x5 devices including 935, 955, and 975
have moved to em2828x usb bridge. Support for the
em2828x usb bridge and USB ID's for the new devices
has been submitted.
Signed-off-by: Bradford Love <brad@nextdimension.cc> Signed-off-by: Hans Verkuil <hverkuil+cisco@kernel.org>
Leo Yan [Tue, 17 Mar 2026 18:58:00 +0000 (18:58 +0000)]
tools build: Correct link flags for libopenssl
The perf static build reports that the BPF skeleton is disabled due to
the missing libopenssl feature.
Use PKG_CONFIG to determine the link flags for libopenssl. Add
"--static" to the PKG_CONFIG command for static linking.
Fixes: 7678523109d1 ("tools/build: Add a feature test for libopenssl") Signed-off-by: Leo Yan <leo.yan@arm.com> Signed-off-by: Namhyung Kim <namhyung@kernel.org>
These guards appear to be leftovers from file renames or copy-paste
errors. Correcting them to follow the standard convention of matching
the file name prevents potential include guard collisions.
Douglas Anderson [Tue, 17 Mar 2026 16:01:20 +0000 (09:01 -0700)]
device property: Make modifications of fwnode "flags" thread safe
In various places in the kernel, we modify the fwnode "flags" member
by doing either:
fwnode->flags |= SOME_FLAG;
fwnode->flags &= ~SOME_FLAG;
This type of modification is not thread-safe. If two threads are both
mucking with the flags at the same time then one can clobber the
other.
While flags are often modified while under the "fwnode_link_lock",
this is not universally true.
Create some accessor functions for setting, clearing, and testing the
FWNODE flags and move all users to these accessor functions. New
accessor functions use set_bit() and clear_bit(), which are
thread-safe.
Cc: stable@vger.kernel.org Fixes: c2c724c868c4 ("driver core: Add fw_devlink_parse_fwtree()") Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> Acked-by: Mark Brown <broonie@kernel.org> Reviewed-by: Wolfram Sang <wsa+renesas@sang-engineering.com> Signed-off-by: Douglas Anderson <dianders@chromium.org> Reviewed-by: Rafael J. Wysocki (Intel) <rafael@kernel.org> Reviewed-by: Saravana Kannan <saravanak@kernel.org> Link: https://patch.msgid.link/20260317090112.v2.1.I0a4d03104ecd5103df3d76f66c8d21b1d15a2e38@changeid
[ Fix fwnode_clear_flag() argument alignment, restore dropped blank
line in fwnode_dev_initialized(), and remove unnecessary parentheses
around fwnode_test_flag() calls. - Danilo ] Signed-off-by: Danilo Krummrich <dakr@kernel.org>
Andrii Nakryiko [Thu, 26 Mar 2026 20:53:57 +0000 (13:53 -0700)]
Merge branch 'add-btf-layout-to-btf'
Alan Maguire says:
====================
Add BTF layout to BTF
Update struct btf_header to add a new "layout" section containing
a description of how to parse the BTF kinds known about at BTF
encoding time. This provides the opportunity for tools that might
not know all of these kinds - as is the case when older tools run
on more newly-generated BTF - to still parse the BTF provided,
even if it cannot all be used.
The ideas here were discussed at [1], with further discussion
at [2].
Patches for pahole will enable the layout addition during BTF
generation are at [3], but even absent these the addition of the
layout feature in the final patch in this series should not break
anything since such unknown features are simply ignored during pahole
BTF generation.
Separately tested sanitization of BTF location info with separate
small series which simulates feature absence to support testing of
features for older kernels; will follow up with that shortly.
Changes since v15 [4]:
- Fixed endian issues for layout section by swapping flags fields
where needed (sashiko.dev, patch 2)
- Fixed string size issue with swapped endian case, use btf->magic
for comparison to determine endian mismatch (bpf review bot,
sashiko.dev, patch 6)
Changes since v14 [5]:
- Fix potential overflow for swapped endian case (BPF review bot,
patch 2)
- Add global: keyword to libbpf.map (sashiko.dev, patch 4)
- Fix endian issues in sanitization; we use the endian safe
btf->hdr and check for endian mismatch between it and raw original
BTF header to inform how we write the change str_off. Also fix
potential truncation issues due to not including hdr->type_off
(sashiko.dev, patch 6)
- Fix issues with selftests raw BTF file interactions (sashiko.dev,
patch 8)
- Drop feature test test since it will be covered by another series
Changes since v13: [6]:
- add feature check/sanitization of BTF with layout info (Andrii,
patch 6)
- added feature check test for layout support (patch 9)
Changes since v12: [7]:
- add logging of layout off/len to kernel header logging (review bot,
patch 6)
- add mode to open() in selftest (review bot, patch 7)
Changes since v11 [8]:
- Revert unneeded changes to btf__new_empty() (Eduard, review bot,
patch 4)
- Reorder btf_parse_layout_sec() checks to ensure min size check
occurs before multiple check (review bot, patch 6)
Changes since v10 [9]:
- deal with read overflow with small header (review bot, patch 2)
- validate layout length is a multiple of sizeof(struct btf_layout)
(review bot, patch 6)
- fix comment style (Alexei, patches 4,7)
- remove bpftool BTF metadata subcommands for now (Alexei)
Changes since v9: [10]:
- fix memcpy header size overrun (review bot, patch 2)
- return size computation directly (Andrii, patch 333)
- revert to original unknown kind logging (Alexei/review bot,
patch 6)
- gap-checking logic can be simplified now that we have
4-byte aligned types and layout together (patch 6)
- fix naming of layout offset, unconditionally emit a layouts
array in json (Quentin, review bot, patch 8)
- fix metadata output in man page to include flags (review bot,
patch 9)
Changes since v8: [11]:
- updated name from "kind_layout" to "layout" (Andrii)
- moved layout info to inbetween types and strings since
both types and layout info align on 4 bytes (Andrii)
- use embedded btf_header (Eduard)
- when consulting layout, fall back to base BTF if none found in
split BTF; this will allow us to only encode layout info in
vmlinux rather than repeating it for each module.
Changes since v7: [12]:
- Fixed comment style in UAPI headers (Mykyta, patch 1)
- Simplify calcuation of header size using min() (Mykyta, patch 2)
- simplify computation of bounds for kind (Mykyta, patch 3)
- Added utility functions for updating type, string offsets when
data is added; this simplifies the code and encapsulates such
updates more clearly (patch 2)
Changes since v6: [13]:
- BPF review bot caught some memory leaks around freeing
of kind layout; more importantly, it noted that we were
breaking with the contiguous BTF representation for
btf_new_empty_opts(). Doing so meant that freeing kind_layout
could not be predicated on having btf->modifiable set, so
adpoted the contiguous raw data layout for BTF to be
consistent with type/string storage (patches 2,4)
- Moved checks for kind overflow prior to referencing kinds
to avoid any risk of overrun (patches 3, 8)
- Tightened up kind layout header offset/len header validation
to catch invalid combinations early in btf_parse_hdr()
(patch 2)
- Fixed selftest to verify calloc success (patch 7)
Changes since v5: [14]:
- removed flags field from kind layout; it is not really workable
since we would have to define semantics of all possible future
flags today to be usable. Instead stick to parsing only, which
means each kind just needs the length of the singular and
vlen-specified objects (Alexei)
- added documentation for bpftool BTF metadata dump (Quentin, patch 9)
Changes since v4: [15]:
- removed CRC generation since it is not needed to handle modules
built at different time than kernel; distilled base BTF supports
this now
- fixed up bpftool display of empty kind names, comment/documentation
indentation (Quentin, patches 8, 9)
Changes since v3 [16]:
- fixed mismerge issues with kbuild changes for BTF generation
(patches 9, 14)
- fixed a few small issues in libbpf with kind layout representation
(patches 2, 4)
Changes since v2 [17]:
- drop "optional" kind flag (Andrii, patch 1)
- allocate "struct btf_header" for struct btf to ensure
we can always access new fields (Andrii, patch 2)
- use an internal BTF kind array in btf.c to simplify
kind encoding (Andrii, patch 2)
- drop use of kind layout information for in-kernel parsing,
since the kernel needs to be strict in what it accepts
(Andrii, patch 6)
- added CRC verification for BTF objects and for matching
with base object (Alexei, patches 7,8)
- fixed bpftool json output (Quentin, patch 10)
- added standalone module BTF support, tests (patches 13-17)
Changes since RFC
- Terminology change from meta -> kind_layout
(Alexei and Andrii)
- Simplify representation, removing meta header
and just having kind layout section (Alexei)
- Fixed bpftool to have JSON support, support
prefix match, documented changes (Quentin)
- Separated metadata opts into add_kind_layout
and add_crc
- Added additional positive/negative tests
to cover basic unknown kind, one with an
info_sz object following it and one with
N elem_sz elements following it.
- Updated pahole-flags to use help output
rather than version to see if features
are present
Alan Maguire [Thu, 26 Mar 2026 14:54:44 +0000 (14:54 +0000)]
kbuild, bpf: Specify "layout" optional feature
The "layout" feature will add metadata about BTF kinds to the
generated BTF; its absence in pahole will not trigger an error so it
is safe to add unconditionally as it will simply be ignored if pahole
does not support it.
Alan Maguire [Thu, 26 Mar 2026 14:54:43 +0000 (14:54 +0000)]
selftests/bpf: Test kind encoding/decoding
verify btf__new_empty_opts() adds layouts for all kinds supported,
and after adding kind-related types for an unknown kind, ensure that
parsing uses this info when that kind is encountered rather than
giving up.
Alan Maguire [Thu, 26 Mar 2026 14:54:42 +0000 (14:54 +0000)]
btf: Support kernel parsing of BTF with layout info
Validate layout if present, but because the kernel must be
strict in what it accepts, reject BTF with unsupported kinds,
even if they are in the layout information.
Alan Maguire [Thu, 26 Mar 2026 14:54:41 +0000 (14:54 +0000)]
libbpf: Support sanitization of BTF layout for older kernels
Add a FEAT_BTF_LAYOUT feature check which checks if the
kernel supports BTF layout information. Also sanitize
BTF if it contains layout data but the kernel does not
support it. The sanitization requires rewriting raw
BTF data to update the header and eliminate the layout
section (since it lies between the types and strings),
so refactor sanitization to do the raw BTF retrieval
and creation of updated BTF, returning that new BTF
on success.
Alan Maguire [Thu, 26 Mar 2026 14:54:40 +0000 (14:54 +0000)]
libbpf: BTF validation can use layout for unknown kinds
BTF parsing can use layout to navigate unknown kinds, so
btf_validate_type() should take layout information into
account to avoid failure when an unrecognized kind is met.
Alan Maguire [Thu, 26 Mar 2026 14:54:39 +0000 (14:54 +0000)]
libbpf: Add layout encoding support
Support encoding of BTF layout data via btf__new_empty_opts().
Current supported opts are base_btf and add_layout.
Layout information is maintained in btf.c in the layouts[] array;
when BTF is created with the add_layout option it represents the
current view of supported BTF kinds.
Alan Maguire [Thu, 26 Mar 2026 14:54:37 +0000 (14:54 +0000)]
libbpf: Support layout section handling in BTF
Support reading in layout fixing endian issues on reading;
also support writing layout section to raw BTF object.
There is not yet an API to populate the layout with meaningful
information.
As part of this, we need to consider multiple valid BTF header
sizes; the original or the layout-extended headers.
So to support this, the "struct btf" representation is modified
to contain a "struct btf_header" and we copy the valid
portion from the raw data to it; this means we can always safely
check fields like btf->hdr.layout_len .
Note if parsed-in BTF has extra header information beyond
sizeof(struct btf_header) - if so we make that BTF ineligible
for modification by setting btf->has_hdr_extra .
Ensure that we handle endianness issues for BTF layout section,
though currently only field that needs this (flags) is unused.
Alan Maguire [Thu, 26 Mar 2026 14:54:36 +0000 (14:54 +0000)]
btf: Add BTF kind layout encoding to UAPI
BTF kind layouts provide information to parse BTF kinds. By separating
parsing BTF from using all the information it provides, we allow BTF
to encode new features even if they cannot be used by readers. This
will be helpful in particular for cases where older tools are used
to parse newer BTF with kinds the older tools do not recognize;
the BTF can still be parsed in such cases using kind layout.
The intent is to support encoding of kind layouts optionally so that
tools like pahole can add this information. For each kind, we record
- length of singular element following struct btf_type
- length of each of the btf_vlen() elements following
- a (currently unused) flags field
Rodrigo Alencar [Thu, 26 Mar 2026 18:32:16 +0000 (18:32 +0000)]
dt-bindings: iio: amplifiers: ad8366: add adrf5702/3 support
Add compatible entries for ADRF5702 and ADRF5703 Digital Attenuators.
ADRF5702 is an 8-bit DSA with a step of 0.125 dB and ADRF5703 is a 7-bit
DSA with a step 0.25 dB. Then, each device ends up with its own gain
range, hence no fallback compatibles are used.
Reviewed-by: Nuno Sá <nuno.sa@analog.com> Signed-off-by: Rodrigo Alencar <rodrigo.alencar@analog.com> Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
compat.bpf.h defined a fallback SCX_ENQ_IMMED macro using
__COMPAT_ENUM_OR_ZERO(). After 6bf36c68b0a2 ("tools/sched_ext:
Regenerate autogen enum headers") added SCX_ENQ_IMMED to the autogen
headers, including both triggers -Wmacro-redefined warnings.
The autogen definition through const volatile __weak already resolves to
0 on older kernels, providing the same backward compatibility. Remove
the now-redundant compat fallback.
Ilpo Järvinen [Tue, 24 Mar 2026 16:56:24 +0000 (18:56 +0200)]
resource: Add __resource_contains_unbound() for internal contains checks
__find_resource_space() currently uses resource_contains() but for
tentative resources that are not yet crafted into the resource tree. As
resource_contains() checks that IORESOURCE_UNSET is not set for either of
the resources, the caller has to hack around this problem by clearing the
IORESOURCE_UNSET flag (essentially lying to resource_contains()).
Instead of the hack, introduce __resource_contains_unbound() for cases like
this.
Ilpo Järvinen [Fri, 13 Mar 2026 08:45:50 +0000 (10:45 +0200)]
PCI: Fix premature removal from realloc_head list during resource assignment
reassign_resources_sorted() checks for two things:
a) Resource assignment failures for mandatory resources by checking if the
resource remains unassigned, which are known to always repeat, and does
not attempt to assign them again.
b) That resource is not among the ones being processed/assigned at this
stage, leading to skip processing such resources in
reassign_resources_sorted() as well (resource assignment progresses
one PCI hierarchy level at a time).
The problem here is that a) is checked before b), but b) also implies the
resource is not being assigned yet, making also a) true. As a) only skips
resource assignment but still removes the resource from realloc_head, the
later stages that would need to process the information in realloc_head
cannot obtain the optional size information anymore. This leads to
considering only non-optional part for bridge windows deeper in the PCI
hierarchy.
This problem has been observed during rescan (add_size is not considered
while attempting assignment for 0000:e2:00.0 indicating the corresponding
entry was removed from realloc_head while processing resource assignments
for 0000:e1):
Fixes: 96336ec70264 ("PCI: Perform reset_resource() and build fail list in sync") Reported-by: Peter Nisbet <peter.nisbet@intel.com> Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Tested-by: Peter Nisbet <peter.nisbet@intel.com> Link: https://patch.msgid.link/20260313084551.1934-1-ilpo.jarvinen@linux.intel.com
Nick Xie [Wed, 25 Mar 2026 07:06:16 +0000 (15:06 +0800)]
iio: adc: meson-saradc: add support for Meson S4
Add support for the SARADC found on the Amlogic Meson S4 SoC.
According to the documentation and current testing, it is fully
compatible with the G12A parameter set, so we reuse
`meson_sar_adc_g12a_data` for this new compatible string.
Although the device tree fallback mechanism could handle the match,
a dedicated entry is added to ensure the userspace ABI correctly
reports the specific part name ("meson-s4-saradc"). This allows
userspace to accurately identify the exact device and maintains
consistency across different firmware types where automatic fallback
parsing might be problematic.
Reviewed-by: Martin Blumenstingl <martin.blumenstingl@googlemail.com> Signed-off-by: Nick Xie <nick@khadas.com> Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Add the compatible string for the SARADC (Successive Approximation
Register ADC) IP block found in the Amlogic Meson S4 SoC.
There are no known differences between the SARADC on S4 and the one
on G12A. Therefore, it uses "amlogic,meson-g12a-saradc" as a proper
specific fallback.
Also add a comment indicating that "amlogic,meson-saradc" must not be
used for new devices. It's a made up compatible string that does not
correspond to a specific hardware generation and is not used to match
any driver. For old devices we keep it as it's part of the ABI.
Signed-off-by: Nick Xie <nick@khadas.com> Reviewed-by: Krzysztof Kozlowski <krzysztof.kozlowski@oss.qualcomm.com> Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Milan Misic [Tue, 24 Mar 2026 19:36:26 +0000 (20:36 +0100)]
iio: imu: st_lsm6dsx: Add ACPI ID for SHIFT13mi gyroscope
The SHIFT13mi or SHIFTbook tablet device by the German manufacturer
SHIFT contains an STM LSM6DSO IMU declared in the DSDT with the
hardware ID SMOCF00. Add this ID to the ACPI match table so that the
driver binds correctly to this device.
Signed-off-by: Milan Misic <twoexem@gmail.com> Reviewed-by: Andy Shevchenko <andriy.shevchenko@intel.com> Signed-off-by: Jonathan Cameron <Jonathan.Cameron@huawei.com>
Linus Torvalds [Thu, 26 Mar 2026 19:42:28 +0000 (12:42 -0700)]
Merge tag 'pm-7.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull power management fixes from Rafael Wysocki:
"These fix two cpufreq issues, one in the core and one in the
conservative governor, and two issues related to system sleep:
- Restore the cpufreq core behavior changed inadvertently during the
6.19 development cycle to call cpufreq_frequency_table_cpuinfo()
for cpufreq policies getting re-initialized which ensures that
policy->max and policy->cpuinfo_max_freq will be valid going
forward (Viresh Kumar)
- Adjust the cached requested frequency in the conservative cpufreq
governor on policy limits changes to prevent it from becoming stale
in some cases (Viresh Kumar)
- Prevent pm_restore_gfp_mask() from triggering a WARN_ON() in some
code paths in which it is legitimately called without invoking
pm_restrict_gfp_mask() previously (Youngjun Park)
- Update snapshot_write_finalize() to take trailing zero pages into
account properly which prevents user space restore from failing
subsequently in some cases (Alberto Garcia)"
* tag 'pm-7.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
PM: sleep: Drop spurious WARN_ON() from pm_restore_gfp_mask()
PM: hibernate: Drain trailing zero pages on userspace restore
cpufreq: conservative: Reset requested_freq on limits change
cpufreq: Don't skip cpufreq_frequency_table_cpuinfo()
Ilpo Järvinen [Thu, 19 Feb 2026 15:39:51 +0000 (17:39 +0200)]
PCI: Prevent shrinking bridge window from its required size
Steve reported an eGPU (either Radeon Instinct MI50 32GB or NVIDIA 3080
10GB) connected via Thunderbolt was not assigned sufficient BAR space in
v6.11, so the amdgpu and nvidia drivers were unable to initialize the
device.
pci_bridge_distribute_available_resources() -> ... ->
adjust_bridge_window() is called between __pci_bus_size_bridges()
and assigning the resources. Since the commit 948675736a77 ("PCI: Allow
adjust_bridge_window() to shrink resource if necessary")
adjust_bridge_window() can also shrink the bridge window. The shrunken
size, however, conflicts with what __pci_bus_size_bridges() ->
pbus_size_mem() calculated as the required bridge window size. By shrinking
the size, adjust_bridge_window() prevents the rest of the resource fitting
algorithm from working as intended. Resource fitting logic is expecting
assignment failures when bridge windows need resizing, but there are cases
where failures are no longer happening after the commit 948675736a77 ("PCI:
Allow adjust_bridge_window() to shrink resource if necessary").
The commit 948675736a77 ("PCI: Allow adjust_bridge_window() to shrink
resource if necessary") justifies the change by the extra reservation
made due to hpmemsize parameter, however, the kernel code contradicts
that statement. (For simplicity, finer-grained hpmmiosize and hpmmiopref
parameters that can be used to the same effect as hpmemsize are ignored in
this description.)
pbus_size_mem() calls calculate_memsize() twice. First with add_size=0
to find out the minimal required resource size. The second call occurs
with add_size=hpmemsize (effectively) but the result does not directly
affect the resource size only resulting in an entry on the realloc_head
list (a.k.a. add_list). Yet, adjust_bridge_window() directly changes
the resource size which does not include what is reserved due to
hpmemsize. Also, if the required size for the bridge window exceeds
hpmemsize, the parameter does not have any effect even on the second
size calculation made by pbus_size_mem(); from calculate_memsize():
size = max(size, add_size) + children_add_size;
The commit ae4611f1d7e9 ("PCI: Set resource size directly in
adjust_bridge_window()") that precedes the commit 948675736a77 ("PCI:
Allow adjust_bridge_window() to shrink resource if necessary") is also
related to causing this problem. Its changelog explicitly states
adjust_bridge_window() wants to "guarantee" allocation success.
Guaranteed allocations, however, are incompatible with how the other
parts of the resource fitting algorithm work. The given justification
fails to explain why guaranteed allocations at this stage are required
nor why forcing window to a smaller value than what was calculated by
pbus_size_mem() is correct. While the change might have worked by chance
in some test scenario, too small bridge window does not "guarantee"
success from the point of view of the endpoint device resource
assignments. No issue is mentioned within the changelog so it's unclear
if the change was made to fix some observed issue nor and what that
issue was.
The unwanted shrinking of a bridge window occurs, e.g., when a device with
large BARs such as eGPU is attached using Thunderbolt and the Root Port
holds less than enough resource space for the eGPU. The GPU resources are
in order of GBs and the default hotplug allocation is a mere 2MB
(DEFAULT_HOTPLUG_MMIO_PREF_SIZE). The problem is illustrated by this log
(filtered to the relevant content only):
The initial size of the Root Port's window is 448MB (0x601bffffff -
0x6000000000). __pci_bus_size_bridges() -> pbus_size_mem() calculates the
required size to be 32772 MB (0x10003fffff - 0x800000000) which would fit
the eGPU resources. adjust_bridge_window() then shrinks the bridge window
down to what is guaranteed to fit into the Root Port's bridge window. The
bridge window for 03:00.0 is also eliminated from the add_list (a.k.a.
realloc_head) list by adjust_bridge_window().
After adjustment, the resources are assigned and as the bridge window for
03:00.0 is assigned successfully, no failure is recorded. Without a
failure, no attempt to resize the window of the Root Port is required. The
end result is eGPU not having large enough resources to work.
The commit 948675736a77 ("PCI: Allow adjust_bridge_window() to shrink
resource if necessary") also claims nested bridge windows are sized the
same, which is false. pbus_size_mem() calculates the size for the parent
bridge window by summing all the downstream resources so the resource
fitting calculates larger bridge window for the parent to accommodate the
childen. That is, hpmemsize does not result the same size for the case
where there are nested bridge windows.
In order to fix the most immediate problem, don't shrink the resource size
in adjust_bridge_window() as hpmemsize had nothing to do with it. When
considering add_size, only reduce it up to what is added due to hpmemsize
(if required size is larger than hpmemsize, the parameter has no impact,
see calculate_memsize()). Unfortunately, if the tail of the bridge window
was aligned in calculate_memsize() from below hpmemsize to above it, the
size check will falsely match but the check at least errs to the side of
caution. There's not enough information available in adjust_bridge_window()
to know the calculated size precisely.
This is not exactly a revert of the commits e4611f1d7e9 ("PCI: Set resource
size directly in adjust_bridge_window()") and 948675736a77 ("PCI: Allow
adjust_bridge_window() to shrink resource if necessary") as shrinking still
remains in place but is implemented differently, and the end result behaves
very differently.
It is possible that those two commits fixed some other issue that is not
described with enough detail in the changelog and undoing parts of them
results in another regression due to behavioral change. Nonetheless, as
described above, the solution by those two commits was flawed and the
issue, if one exists, should be solved in a way that is compatible with the
rest of the resource fitting algorithm instead of working against it.
Besides shrinking, the case where adjust_bridge_window() expands the bridge
window is likely somewhat wrong as well because it removes the entry from
add_list (a.k.a. realloc_head), but it is less damaging as that only
impacts optional resources and may have no impact if expanding by hpmemsize
is larger than what add_size was. Fixing it is left as further work.
Fixes: 948675736a77 ("PCI: Allow adjust_bridge_window() to shrink resource if necessary") Fixes: ae4611f1d7e9 ("PCI: Set resource size directly in adjust_bridge_window()") Reported-by: Steve Oswald <stevepeter.oswald@gmail.com> Closes: https://lore.kernel.org/linux-pci/CAN95MYEaO8QYYL=5cN19nv_qDGuuP5QOD17pD_ed6a7UqFVZ-g@mail.gmail.com/ Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@linux.intel.com> Signed-off-by: Bjorn Helgaas <bhelgaas@google.com> Link: https://patch.msgid.link/20260219153951.68869-1-ilpo.jarvinen@linux.intel.com
Ahmed Naseef [Thu, 12 Mar 2026 16:53:32 +0000 (16:53 +0000)]
PCI: Prevent assignment to unsupported bridge windows
Previously, pci_read_bridge_io() and pci_read_bridge_mmio_pref()
unconditionally set resource type flags (IORESOURCE_IO or IORESOURCE_MEM |
IORESOURCE_PREFETCH) when reading bridge window registers. For windows that
are not implemented in hardware, this may cause the allocator to assign
space for a window that doesn't exist.
For example, the EcoNET EN7528 SoC Root Port doesn't support the
prefetchable window, but since a downstream device had a prefetchable BAR,
the allocator mistakenly assigned a prefetchable window:
pci 0001:00:01.0: [14c3:0811] type 01 class 0x060400 PCIe Root Port
pci 0001:00:01.0: PCI bridge to [bus 01-ff]
pci 0001:00:01.0: bridge window [mem 0x28000000-0x280fffff]: assigned
pci 0001:00:01.0: bridge window [mem 0x28100000-0x282fffff pref]: assigned
pci 0001:01:00.0: BAR 0 [mem 0x28100000-0x281fffff 64bit pref]: assigned
pci_read_bridge_windows() already detects unsupported windows by testing
register writability and sets dev->io_window/pref_window accordingly.
Check dev->io_window/pref_window so we don't set the resource flags for
unsupported windows, which prevents the allocator from assigning space to
them.
After this commit, the prefetchable BAR is correctly allocated from the
non-prefetchable window:
Linus Torvalds [Thu, 26 Mar 2026 19:27:17 +0000 (12:27 -0700)]
Merge tag 'thermal-7.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull thermal control fix from Rafael Wysocki:
"This prevents the int340x thermal driver from taking the power slider
offset parameter into account incorrectly in some cases (Srinivas
Pandruvada)"
* tag 'thermal-7.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
thermal: intel: int340x: soc_slider: Set offset only for balanced mode
Abel Vesa [Fri, 13 Mar 2026 12:22:38 +0000 (14:22 +0200)]
dt-bindings: spmi: qcom,x1e80100-spmi-pmic-arb: Document Eliza compatible
The SPMI multi-master Arbiter found on Eliza is version 7.2.0, yet
driver-wise, still compatible with the one featured on Hamoa (X1E80100),
which is 7.0.1.
So document the Eliza compatible and allow Hamoa one as fallback.
Move the content of fdt_reserved_mem_save_node() to
fdt_init_reserved_mem_node() function. Initialization is no longer
performed in two steps as it was initially, so
fdt_reserved_mem_save_node() name is a bit misleading and that function
now performs full initialization of the reserved memory region.
This also fixes the problem of keeping pointers to the regions, which
failed to initialize, what might cause issues when such region is
assigned to the device.
Rename fdt_scan_reserved_mem_reg_nodes() to fdt_scan_reserved_mem_late()
to clearly show how it differs from fdt_scan_reserved_mem() and update
description of both functions.
Marek Szyprowski [Wed, 25 Mar 2026 09:00:21 +0000 (10:00 +0100)]
of: reserved_mem: rearrange code a bit
Move __rmem_check_for_overlap() and __rmem_cmp() functions before
fdt_scan_reserved_mem_reg_nodes() to avoid forward declaration and keep
related code close together.
Marek Szyprowski [Wed, 25 Mar 2026 09:00:19 +0000 (10:00 +0100)]
of: reserved_mem: switch to ops based OF_DECLARE()
Move init function from OF_DECLARE() argument to the given reserved
memory region ops structure and then pass that structure to the
OF_DECLARE() initializer. This node_init callback is mandatory for the
reserved mem driver. Such change makes it possible in the future to add
more functions called by the generic code before given memory region is
initialized and rmem object is created.
Marek Szyprowski [Wed, 25 Mar 2026 09:00:18 +0000 (10:00 +0100)]
of: reserved_mem: use -ENODEV instead of -ENOENT
When given reserved memory region doesn't really support given node,
return -ENODEV instead of -ENOENT. Then fix __reserved_mem_init_node()
function to properly propagate error code different from -ENODEV instead
of silently ignoring it.
Linus Torvalds [Thu, 26 Mar 2026 19:06:40 +0000 (12:06 -0700)]
Merge tag 'acpi-7.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm
Pull ACPI support fix from Rafael Wysocki:
"Prevent use-after-free from occurring on reduced-hardware ACPI
platforms when -EPROBE_DEFER is returned by ec_install_handlers()
during ACPI EC driver initialization (Weiming Shi)"
* tag 'acpi-7.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/rafael/linux-pm:
ACPI: EC: clean up handlers on probe failure in acpi_ec_setup()
Biju Das [Thu, 26 Mar 2026 16:24:51 +0000 (16:24 +0000)]
pinctrl: renesas: rzg2l: Fix save/restore of {IOLH,IEN,PUPD,SMT} registers
The rzg2l_pinctrl_pm_setup_regs() handles save/restore of
{IOLH,IEN,PUPD,SMT} registers during s2ram, but only for ports where all
pins share the same pincfg. Extend the code to also support ports with
variable pincfg per pin, so that {IOLH,IEN,PUPD,SMT} registers are
correctly saved and restored for all pins.
Lad Prabhakar [Thu, 19 Mar 2026 14:15:15 +0000 (14:15 +0000)]
pinctrl: renesas: rzt2h: Add pin configuration support
Add pin configuration support for the Renesas RZ/T2H SoC. The RZ/T2H SoC
allows configuring several electrical characteristics through the DRCTLm
(I/O Buffer Function Switching) registers. These registers control bias
configuration, Schmitt trigger input, output slew rate, and drive
strength.
Implement pinconf_ops to allow reading and updating these properties
through the generic pin configuration framework. The implementation
supports bias-disable, bias-pull-up, bias-pull-down,
input-schmitt-enable, slew-rate, and drive-strength-microamp.
Linus Torvalds [Thu, 26 Mar 2026 19:03:37 +0000 (12:03 -0700)]
Merge tag 'landlock-7.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/mic/linux
Pull Landlock fixes from Mickaël Salaün:
"This mainly fixes Landlock TSYNC issues related to interrupts and
unexpected task exit.
Other fixes touch documentation and sample, and a new test extends
coverage"
* tag 'landlock-7.0-rc6' of git://git.kernel.org/pub/scm/linux/kernel/git/mic/linux:
landlock: Expand restrict flags example for ABI version 8
selftests/landlock: Test tsync interruption and cancellation paths
landlock: Clean up interrupted thread logic in TSYNC
landlock: Serialize TSYNC thread restriction
samples/landlock: Bump ABI version to 8
landlock: Improve TSYNC types
landlock: Fully release unused TSYNC work entries
landlock: Fix formatting
Document the pin configuration properties supported by the RZ/T2H
pinctrl driver.
The RZ/T2H SoC allows configuring several electrical characteristics
through the DRCTLm (I/O Buffer Function Switching) registers. These
registers control drive strength, bias configuration, Schmitt trigger
input, and output slew rate.
Matthew Brost [Thu, 26 Mar 2026 01:12:07 +0000 (18:12 -0700)]
drm/xe: Issue GGTT invalidation under lock in ggtt_node_remove
Async work (e.g., GuC queue teardowns) can call ggtt_node_remove, so the
operation must be performed under the GGTT lock to ensure the GGTT
online check remains stable. GGTT insertion and removal are heavyweight
operations (e.g., queue create/destroy), so the additional serialization
cost is negligible compared to ensuring correctness.
Marek Vasut [Thu, 26 Mar 2026 04:54:59 +0000 (05:54 +0100)]
ARM: dts: renesas: Drop KSZ8041 PHY C22 compatible strings
The Microchip KSZ8041 PHY schema indicates that the compatible string
"ethernet-phy-id0022.1537" must not be followed by any other compatible
string. Drop trailing "ethernet-phy-ieee802.3-c22" to match the schema.
Marek Vasut [Thu, 26 Mar 2026 04:54:01 +0000 (05:54 +0100)]
ARM: dts: renesas: rza2mevb: Drop RTL8201F PHY C22 compatible string
The Realtek RTL8201F PHY schema indicates that the compatible string
"ethernet-phy-id001c.c816" must not be followed by any other compatible
string. Drop trailing "ethernet-phy-ieee802.3-c22" to match the schema.
Marek Vasut [Thu, 26 Mar 2026 04:53:38 +0000 (05:53 +0100)]
ARM: dts: renesas: r8a7742-iwg21d-q7-dbcm-ca: Drop KSZ8081 PHY C22 compatible string
The Microchip KSZ8081 PHY schema indicates that the compatible string
"ethernet-phy-id0022.1560" must not be followed by any other compatible
string. Drop trailing "ethernet-phy-ieee802.3-c22" to match the schema.
Biju Das [Tue, 24 Mar 2026 11:43:13 +0000 (11:43 +0000)]
arm64: dts: renesas: Add initial DTSI for RZ/G3L SoC
Add the initial DTSI for the RZ/G3L SoC.
The files in this commit have the following meaning:
- r9a08g046.dtsi: RZ/G3L family SoC common parts
- r9a08g046l48.dtsi: RZ/G3L R9A08G046L48 SoC-specific parts
Add placeholders to reuse the code for the Renesas SMARC II carrier
board.
Document the device tree bindings for the Renesas RZ/G3L SoC Clock Pulse
Generator (CPG). RZ/G3L CPG is similar to RZ/G2L CPG but has 5 clocks
compared to 1 clock on other SoCs.
Also define RZ/G3L (R9A08G046) Clock Pulse Generator Core Clocks, as
listed in section 4.4.4.1 ("Block Diagram of the Clock System"), module
clock outputs, as listed in section 4.4.2 ("Clock List r1.00") and add
Reset definitions referring to registers CPG_RST_* in Section 4.4.3
("Register") of the RZ/G3L Hardware User's Manual (Rev.1.00 Oct, 2025).
Biju Das [Tue, 24 Mar 2026 11:43:11 +0000 (11:43 +0000)]
clk: renesas: rzg2l: Re-enable critical module clocks during resume
After a suspend/resume cycle, critical module clocks (CLK_IS_CRITICAL)
may be left disabled as there is no owning driver to restore them,
unlike regular clocks.
Add rzg2l_mod_enable_crit_clock_init_mstop() which walks all module
clocks on resume, re-enables any critical clock found disabled, and then
restores the MSTOP state for clocks that have one via the existing
helper. This replaces the direct call to rzg2l_mod_clock_init_mstop()
in rzg2l_cpg_resume(), preserving the correct clock-before-MSTOP restore
ordering.
Refactor the mstop initialisation logic in rzg2l_mod_clock_init_mstop()
into a dedicated helper function rzg2l_mod_clock_init_mstop_helper().
This decouples the logic for setting module stop state on disabled
clocks from the iteration loop, allowing it to be reused during resume
to re-enable critical clocks.
Biju Das [Tue, 24 Mar 2026 11:43:09 +0000 (11:43 +0000)]
clk: renesas: rzg2l: Add helper for mod clock enable/disable
Refactor rzg2l_mod_clock_endisable() by extracting its logic into a new
helper function rzg2l_mod_clock_endisable_helper(), which accepts an
additional set_mstop_state boolean parameter. This allows callers to
control whether the module stop state is updated alongside the clock
enable/disable operation. No functional change for existing callers.
The RZ/G2L SoC family requires DMA resets to be deasserted for routing
some peripheral interrupts to the CPU. Asserting these resets after boot
would silently break interrupt delivery with no driver to restore them.
Mark the DMA resets as critical by adding them to the crit_resets table
in the SoC-specific rzg2l_cpg_info for r9a07g043, r9a07g044, and
r9a08g045, preventing __rzg2l_cpg_assert() from asserting them and
ensuring they are deasserted during probe and resume.
Biju Das [Tue, 24 Mar 2026 11:43:07 +0000 (11:43 +0000)]
clk: renesas: rzg2l: Add support for critical resets
Some reset lines must remain deasserted at all times after boot, as
asserting them would disable critical system functionality with no
owning driver to restore them. This mirrors the existing crit_mod_clks
mechanism which protects critical module clocks from being disabled.
On RZ/G2L family SoCs, the DMA reset must be remain deasserted for
routing some peripheral interrupts to CPU.
Add crit_resets and num_crit_resets fields to struct rzg2l_cpg_info to
allow SoC-specific data tables to declare reset IDs that must never be
asserted.
Introduce rzg2l_cpg_deassert_crit_resets() to iterate over all critical
resets and deassert them. Call it both at probe time and during resume
to ensure critical peripherals are held out of reset after power-on and
suspend/resume cycles.
The of_pci_get_max_link_speed() function currently validates the
"max-link-speed" DT property to be in the range 1..4 (Gen1..Gen4).
This imposes a maintenance burden because each new PCIe generation
would require updating this validation.
Remove the range check so the function returns the raw property value
(or a negative error code if the property is missing or malformed).
Since the callers are now validating the returned speed against the
range they support, this check can now be safely removed.
Removing the validation from this common function also allows future PCIe
generations to be supported without modifying drivers/pci/of.c.
Hans Zhang [Fri, 13 Mar 2026 16:55:21 +0000 (00:55 +0800)]
PCI: controller: Validate max-link-speed
Add validation for the "max-link-speed" DT property in three more
drivers, using the pcie_get_link_speed() helper.
- brcmstb: If the value is missing or invalid, fall back to no
limitation (pcie->gen = 0). Fix the previous incorrect logic.
- mediatek-gen3: If the value is missing or invalid, use the maximum
speed supported by the controller.
- rzg3s-host: If the value is missing or invalid, fall back to Gen2.
This ensures that all users of of_pci_get_max_link_speed() are ready
for the removal of the central range check.
Hans Zhang [Fri, 13 Mar 2026 16:55:20 +0000 (00:55 +0800)]
PCI: j721e: Validate max-link-speed from DT
Use the new pcie_get_link_speed() helper to validate the value read from
the "max-link-speed" DT property. If the value is missing or invalid,
fall back to Gen2 (speed = 2). This prepares for the removal of the
range check in of_pci_get_max_link_speed().
Hans Zhang [Fri, 13 Mar 2026 16:55:19 +0000 (00:55 +0800)]
PCI: dwc: Use pcie_get_link_speed() helper for safe array access
Replace direct indexing of pcie_link_speed[] with the new helper
pcie_get_link_speed() in all DesignWare core and glue drivers. This
ensures that out-of-range generation numbers do not cause out-of-bounds
accesses when the helper returns PCI_SPEED_UNKNOWN, and prepares for
the removal of the range check in of_pci_get_max_link_speed().
The actual validation of the "max-link-speed" DT property (e.g., fallback
to a safe default and warning) is added in subsequent patches for each
driver that reads the property.
Hans Zhang [Fri, 13 Mar 2026 16:55:18 +0000 (00:55 +0800)]
PCI: Add pcie_get_link_speed() helper for safe array access
The pcie_link_speed[] array is indexed by PCIe generation numbers
(1 = 2.5 GT/s, 2 = 5 GT/s, ...). Several drivers use it directly,
which can lead to out-of-bounds accesses if an invalid generation
number is used.
Introduce a helper function pcie_get_link_speed() that returns the
pci_bus_speed value for a given generation number, or PCI_SPEED_UNKNOWN if
the generation is out of range. This will allow us to safely handle
invalid values after the range check is removed from
of_pci_get_max_link_speed().
Yeoreum Yun [Sat, 14 Mar 2026 17:51:26 +0000 (17:51 +0000)]
arm64: cpufeature: Add FEAT_LSUI
Since Armv9.6, FEAT_LSUI introduces atomic instructions that allow
privileged code to access user memory without clearing the PSTATE.PAN
bit. Add CPU feature detection for FEAT_LSUI.
The rest of Qualcomm Interconnect drivers have stopped using
DEFINE_QNODE long ago for the sake of readability. Stop using it inside
the msm8974 interconnect driver too.
Dmitry Baryshkov [Tue, 24 Mar 2026 00:10:43 +0000 (02:10 +0200)]
interconnect: qcom: msm8974: switch to the main icc-rpm driver
In preparation to restoring the ability of MSM8974 driver to work with
the modern kernels, switch the driver to the main icc-rpm set of helper
code.
As platform-specific workarounds, set the get_bw callback (returning 0)
to prevent initial setup from programming INT_MAX into the RPM (which
otherwise might hang the platform) and tell RPM programming code to
ignore -ENXIO errors from the firmware (until the QoS programming is
sorted out).
Dmitry Baryshkov [Tue, 24 Mar 2026 00:10:42 +0000 (02:10 +0200)]
interconnect: qcom: let platforms declare their bugginess
On MSM8974 programming some of the RPM resources results in the
"resource does not exist" messages from the firmware. This occurs even
with the downstream bus driver, which happily ignores the errors. My
assumption is that these resources existed in the earlier firmware
revisions but were later switched to be programmed differently (for the
later platforms corresponding nodes use qos.ap_owned, which prevents
those resources from being programmed.
In preparation for conversion of the MSM8974 driver (which doesn't have
QoS code yet) to the main icc-rpm set of helpers, let the driver declare
that those -ENXIO errors must be ignored (for now). Later, when the QoS
programming is sorted out (and more interconnects are added to the DT),
this quirk might be removed.
Reviewed-by: Brian Masney <bmasney@redhat.com> Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com> Signed-off-by: Dmitry Baryshkov <dmitry.baryshkov@oss.qualcomm.com> Link: https://msgid.link/20260324-msm8974-icc-v2-6-527280043ad8@oss.qualcomm.com Signed-off-by: Georgi Djakov <djakov@kernel.org>
MSM8974 requires a separate get_bw callback, since on that platform
increasing the clock rate for some of the NoCs during boot may lead to
hangs. For the details see commit 9caf2d956cfa ("interconnect: qcom:
msm8974: Don't boost the NoC rate during boot").
Dmitry Baryshkov [Tue, 24 Mar 2026 00:10:39 +0000 (02:10 +0200)]
interconnect: qcom: drop unused is_on flag
The commit 2e2113c8a64f ("interconnect: qcom: rpm: Handle interface
clocks") has added the is_on flag to the qcom_icc_provider, but failed
to actually utilize it. Drop the flag.
Dmitry Baryshkov [Tue, 24 Mar 2026 00:10:38 +0000 (02:10 +0200)]
dt-bindings: interconnect: qcom,msm8974: use qcom,rpm-common
Use qcom,rpm-common schema to declare interconnects property instead
describing it again. In future this will allow the platform to switch to
the two-cell interconnects, adding the tag to the specification.
Hans Zhang [Sun, 15 Mar 2026 15:53:51 +0000 (23:53 +0800)]
PCI: sky1: Use boolean true for is_rc field
The is_rc field in struct cdns_pcie is of type bool. Replace the
integer assignment (1) with the boolean literal true to improve
code clarity and maintain consistency with the type definition.
Dmitry Baryshkov [Tue, 24 Mar 2026 00:10:37 +0000 (02:10 +0200)]
dt-bindings: interconnect: qcom,msm8974: drop bus clocks
Remove the wrong internal RPM bus clock representation that we've been
carrying for years. They are an internal part of the interconnect
fabric. They are not exported by any device and are not supposed to be
used.
PCI: qcom: Advertise Hotplug Slot Capability with no Command Completion support
Qcom PCIe Root Ports advertise hotplug capability in hardware, but do not
support hotplug command completion. As a result, the hotplug commands
issued by the pciehp driver never gets completion notification, leading to
repeated timeout warnings and multi-second delays during boot and
suspend/resume.
Commit a54db86ddc153 ("PCI: qcom: Do not advertise hotplug capability for
IPs v2.7.0 and v1.9.0") mistakenly assumed that the Root Ports doesn't
support Hotplug due to timeouts and disabled the Hotplug functionality
altogether. But the Root Ports does support reporting Hotplug events like
DL_Up/Down events.
So to fix the command completion timeout issues, just set the No Command
Completed Support (NCCS) bit and enable Hotplug in Slot Capability field
back.
Fixes: a54db86ddc153 ("PCI: qcom: Do not advertise hotplug capability for IPs v2.7.0 and v1.9.0") Signed-off-by: Krishna Chaitanya Chundru <krishna.chundru@oss.qualcomm.com>
[mani: renamed function, commit log and added comment] Signed-off-by: Manivannan Sadhasivam <mani@kernel.org> Tested-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com> # Hamoa CRD, tunneled link Reviewed-by: Konrad Dybcio <konrad.dybcio@oss.qualcomm.com> Link: https://patch.msgid.link/20260314-hotplug-v1-1-96ac87d93867@oss.qualcomm.com
- Prevent pm_restore_gfp_mask() from triggering a WARN_ON() in some
code paths in which it is legitimately called without invoking
pm_restrict_gfp_mask() previously (Youngjun Park)
- Update snapshot_write_finalize() to take trailing zero pages into
account properly which prevents user space restore from failing
subsequently in some cases (Alberto Garcia)
* pm-sleep:
PM: sleep: Drop spurious WARN_ON() from pm_restore_gfp_mask()
PM: hibernate: Drain trailing zero pages on userspace restore
Hans Zhang [Sat, 7 Mar 2026 02:01:52 +0000 (10:01 +0800)]
PCI: dwc: Expose PCIe event counters for groups 5 to 7 over debugfs
Extend the DesignWare PCIe controller's debugfs statistical counter
interface with event definitions from groups 5 through 7 as documented
in the DWC PCIe Databook (version 6.30a, section 3.8.2.3, Tables 3-59,
3-60, 3-61). These counters provide visibility into Layer1 non-error
events (link state transitions,ASPM, L1 substates), Layer2 DLLP
exchanges, and Layer3 TLP transactions.
The counters are exposed under the debugfs statistical counter directory,
allowing users to monitor link behavior and diagnose PCIe protocol issues
more effectively.
Bruce Johnston [Tue, 24 Mar 2026 18:06:52 +0000 (14:06 -0400)]
dm vdo: save the formatted metadata to disk
Add vdo_save_super_block() and vdo_save_geometry_block() to perform
asynchronous writes of the super block and geometry block respectively.
Add vdo_clear_layout() to zero the UDS index's first block, the block
map partition, and the recovery journal partition.
These operations are driven by new phases in the pre-load state machine
(PRE_LOAD_PHASE_FORMAT_*), ensuring that disk writes happen during
pre-resume rather than during dmsetup create.
Signed-off-by: Bruce Johnston <bjohnsto@redhat.com> Reviewed-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Bruce Johnston [Tue, 24 Mar 2026 18:06:51 +0000 (14:06 -0400)]
dm vdo: add formatting logic and initialization
Add the core formatting logic. The initialization path is updated to
read the geometry block (block 0 on the storage device). If the block
is entirely zeroed, the device is treated as unformatted and
vdo_format() is called. Otherwise, the existing geometry is parsed
and the VDO is loaded as before.
The vdo_format() function initializes the volume geometry and super
block, and marks the VDO as needing it's layout saved to disk.
Signed-off-by: Bruce Johnston <bjohnsto@redhat.com> Reviewed-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Add vdo_submit_metadata_vio_wait(), a synchronous I/O submission
helper that blocks until completion. This is needed for I/O during
early initialization before work queues are available.
Refactor read_geometry_block() to use it.
Signed-off-by: Bruce Johnston <bjohnsto@redhat.com> Reviewed-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Bruce Johnston [Tue, 24 Mar 2026 18:06:49 +0000 (14:06 -0400)]
dm vdo: add geometry block structure
Introduce a vdo_geometry_block structure, containing a vio and buffer,
mirroring the existing vdo_super_block structure. Both are now
initialized at VDO startup and freed at shutdown, establishing the
infrastructure needed to read and write the geometry block using the
same mechanisms as the super block.
Refactor read_geometry_block() to use the new structure.
Signed-off-by: Bruce Johnston <bjohnsto@redhat.com> Reviewed-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Bruce Johnston [Tue, 24 Mar 2026 18:06:48 +0000 (14:06 -0400)]
dm vdo: add geometry block encoding
Add vdo_encode_volume_geometry() to write the geometry block into a
buffer so that it can be written to disk. The corresponding decode
path already exists.
Signed-off-by: Bruce Johnston <bjohnsto@redhat.com> Reviewed-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
Bruce Johnston [Tue, 24 Mar 2026 18:06:46 +0000 (14:06 -0400)]
dm vdo: add formatting parameters to table line
Extend the dm table line with three new optional parameters:
indexMemory (UDS index memory size), indexSparse (dense vs sparse
index), and slabSize (blocks per allocation slab). These values are
parsed, validated, and stored in the device configuration for use
during formatting.
Rework the slab size constants from the single MAX_VDO_SLAB_BITS into
explicit MIN_VDO_SLAB_BLOCKS, MAX_VDO_SLAB_BLOCKS, and
DEFAULT_VDO_SLAB_BLOCKS values.
Bump the target version from 9.1.0 to 9.2.0 to reflect this table
line change.
Signed-off-by: Bruce Johnston <bjohnsto@redhat.com> Reviewed-by: Matthew Sakai <msakai@redhat.com> Signed-off-by: Mikulas Patocka <mpatocka@redhat.com>
projects / thirdparty / linux.git / log
