The driver's ndo_get_stats64 callback is only reporting mlx5 counters,
without accounting for the netdev stats, causing errors from the network
stack to be invisible in statistics.
Add netdev_stats_to_stats64() call to first populate the counters, then
add mlx5 counters on top, ensuring both are accounted for (where
appropriate).
Fixes: f62b8bb8f2 ("net/mlx5: Extend mlx5_core to support ConnectX-4 Ethernet functionality")
Signed-off-by: Gal Pressman <gal@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/1769411695-18820-4-git-send-email-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
It is possible to unbind the uplink ETH driver while the E-Switch is
in switchdev mode. This leads to netdevice reference counting issues[1],
as the driver removal path was not designed to clean up from this state.
During uplink ETH driver removal (_mlx5e_remove), the code now waits for
any concurrent E-Switch mode transition to finish. It then removes the
REPs auxiliary device, if exists. This ensures a graceful cleanup.
[1]
unregister_netdevice: waiting for eth2 to become free. Usage count = 2
ref_tracker: netdev@00000000c912e04b has 1/1 users at
ib_device_set_netdev+0x130/0x270 [ib_core]
mlx5_ib_vport_rep_load+0xf4/0x3e0 [mlx5_ib]
mlx5_esw_offloads_rep_load+0xc7/0xe0 [mlx5_core]
esw_offloads_enable+0x583/0x900 [mlx5_core]
mlx5_eswitch_enable_locked+0x1b2/0x290 [mlx5_core]
mlx5_devlink_eswitch_mode_set+0x107/0x3e0 [mlx5_core]
devlink_nl_eswitch_set_doit+0x60/0xd0
genl_family_rcv_msg_doit+0xe0/0x130
genl_rcv_msg+0x183/0x290
netlink_rcv_skb+0x4b/0xf0
genl_rcv+0x24/0x40
netlink_unicast+0x255/0x380
netlink_sendmsg+0x1f3/0x420
__sock_sendmsg+0x38/0x60
__sys_sendto+0x119/0x180
__x64_sys_sendto+0x20/0x30
Fixes: 7a9fb35e8c ("net/mlx5e: Do not reload ethernet ports when changing eswitch mode")
Signed-off-by: Shay Drory <shayd@nvidia.com>
Reviewed-by: Mark Bloch <mbloch@nvidia.com>
Signed-off-by: Tariq Toukan <tariqt@nvidia.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/1769411695-18820-2-git-send-email-tariqt@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Some PTP hardware clock (PHC) devices may return -EOPNOTSUPP for
operations like settime, adjtime, or adjfreq. This commonly occurs
with timestamp-only PHC implementations that don't support full clock
control.
For background, syzbot previously exposed a crash risk when PTP clock
drivers lacked required callbacks[1]. Subsequent work[2] made callback
presence a registration requirement. As a result, some drivers (like
iwlwifi MVM/MLD[3]) now provide stub callbacks that return -EOPNOTSUPP
for unsupported operations.
When phc_ctl encounters such devices, the "Operation not supported"
error should be treated as a skip (device limitation) rather than a
test failure. This patch:
- Adds [SKIP] output handling in log_test()
- Detects "Operation not supported" from phc_ctl and returns ksft_skip
- Returns ksft_skip if all tests are skipped, preventing false-positive
results when testing timestamp-only PHC implementations
Link: https://lore.kernel.org/netdev/20251028043216.1971292-1-junjie.cao@intel.com/ [1]
Link: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=dfb073d32cac [2]
Link: https://lore.kernel.org/netdev/20251204123204.9316-1-ziyao@disroot.org/ [3]
Signed-off-by: Junjie Cao <junjie.cao@intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20260126061532.12532-2-junjie.cao@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The kselftest framework defines KSFT_SKIP=4 as the standard exit code
for skipped tests. However, phc.sh currently uses a mix of 'exit 0' and
'exit 1' to indicate skip conditions, which can confuse test harnesses
and CI systems.
This patch introduces ksft_skip=4 variable and unifies all skip exit
paths to use 'exit $ksft_skip', consistent with other selftests like
net/lib.sh and net/fib_nexthops.sh.
Signed-off-by: Junjie Cao <junjie.cao@intel.com>
Reviewed-by: Simon Horman <horms@kernel.org>
Link: https://patch.msgid.link/20260126061532.12532-1-junjie.cao@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The int51x1 driver uses the same requests as USB CDC to handle packet
filtering, but provides its own definitions and function to handle it.
The chip datasheet says the requests are CDC compliant. Replace this
unnecessary code with a reference to usbnet_cdc_update_filter.
Also fix the broken datasheet link and remove an empty comment.
Signed-off-by: Ethan Nelson-Moore <enelsonmoore@gmail.com>
Link: https://patch.msgid.link/20260126044049.40359-1-enelsonmoore@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Add a scheduler that uses BPF arenas to manage task context data.
Signed-off-by: Emil Tsalapatis <emil@etsalapatis.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Add in the scx_userland scheduler that does vruntime-based
scheduling in userspace code and communicates scheduling
decisions to BPF by accessing and modifying globals through
the skeleton.
Cc: Tejun Heo <tj@kernel.org>
Cc: David Vernet <dvernet@meta.com>
Signed-off-by: Emil Tsalapatis <emil@etsalapatis.com>
Signed-off-by: Tejun Heo <tj@kernel.org>
Tony Nguyen says:
====================
Intel Wired LAN Driver Updates 2026-01-26 (ice, idpf)
For ice:
Jake converts ring stats to utilize u64_stats APIs and performs some
cleanups along the way.
Alexander reorganizes layout of Tx and Rx rings for cacheline
locality and utilizes __cacheline_group* macros on the new layouts.
For idpf:
YiFei Zhu adds support for BPF kfunc reporting of hardware Rx timestamps.
* '100GbE' of git://git.kernel.org/pub/scm/linux/kernel/git/tnguy/next-queue:
idpf: export RX hardware timestamping information to XDP
ice: reshuffle and group Rx and Tx queue fields by cachelines
ice: convert all ring stats to u64_stats_t
ice: shorten ring stat names and add accessors
ice: use u64_stats API to access pkts/bytes in dim sample
ice: remove ice_q_stats struct and use struct_group
ice: pass pointer to ice_fetch_u64_stats_per_ring
====================
Link: https://patch.msgid.link/20260126224313.3847849-1-anthony.l.nguyen@intel.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
All drivers that need to report the RX ring count now implement the
get_rx_ring_count callback directly. Remove the legacy fallback path
that obtained this information by calling get_rxnfc with ETHTOOL_GRXRINGS.
This simplifies the code and makes get_rx_ring_count the only way
to retrieve the RX ring count.
Note: ethtool_get_rx_ring_count() returns int to allow returning
-EOPNOTSUPP, while the callback returns u32. The implicit conversion
is safe since RX ring counts will not exceed INT_MAX while we are still
alive.
Signed-off-by: Breno Leitao <leitao@debian.org>
Link: https://patch.msgid.link/20260126-grxring_final-v1-1-0981cb24512e@debian.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Gal Pressman says:
====================
Single MSS length in UDP GSO_PARTIAL
This series addresses an inconsistency in how UDP GSO_PARTIAL handles
the UDP header length field.
Currently, when GSO_PARTIAL segmentation is used, the UDP header length
contains the large MSS size, requiring drivers to manually adjust it
before transmitting. This is inconsistent with how tunnel GSO_PARTIAL
handles outer headers in UDP tunnels, where the length is set to the
single segment size.
This was originally suggested by Alexander Duyck back in 2018:
https://lore.kernel.org/netdev/CAKgT0UcdnUWgr3KQ=RnLKigokkiUuYefmL-ePpDvJOBNpKScFA@mail.gmail.com/
The first patch fixes the core UDP offload code to set the UDP length
field to the single segment size (gso_size + UDP header) instead of the
large MSS size. This provides hardware with a proper template length
value for final segmentation.
The subsequent patches remove the now redundant UDP header length
adjustments from the mlx5e and aquantia drivers, as the core code now
handles this correctly.
I couldn't find any other drivers that support UDP GSO_PARTIAL; idpf
supports UDP segmentation, but it does not use GSO_PARTIAL.
====================
Link: https://patch.msgid.link/20260125121649.778086-1-gal@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
GSO_PARTIAL now takes care of updating the UDP header length, remove the
redundant assignment in aq_nic_map_skb().
Reviewed-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Gal Pressman <gal@nvidia.com>
Link: https://patch.msgid.link/20260125121649.778086-4-gal@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
GSO_PARTIAL now takes care of updating the UDP header length,
mlx5e_udp_gso_handle_tx_skb() is redundant, remove it.
Reviewed-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Tariq Toukan <tariqt@nvidia.com>
Link: https://patch.msgid.link/20260125121649.778086-3-gal@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
In GSO_PARTIAL segmentation, set the UDP length field to the single
segment size (gso_size + UDP header) instead of the large MSS size.
This provides hardware with a template length value for final
segmentation, similar to how tunnel GSO_PARTIAL handles outer headers
in UDP tunnels.
This will remove the need to manually adjust the UDP header length in
the drivers, as can be seen in subsequent patches.
This was suggested by Alex in 2018:
https://lore.kernel.org/netdev/CAKgT0UcdnUWgr3KQ=RnLKigokkiUuYefmL-ePpDvJOBNpKScFA@mail.gmail.com/
Reviewed-by: Dragos Tatulea <dtatulea@nvidia.com>
Signed-off-by: Gal Pressman <gal@nvidia.com>
Reviewed-by: Willem de Bruijn <willemb@google.com>
Link: https://patch.msgid.link/20260125121649.778086-2-gal@nvidia.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
As the stmmac_priv struct is passed to the fix_soc_reset() method which
has the ioaddr, there is no need to pass ioaddr separately. Pass just
the stmmac_priv struct. Fix up the glues that use it.
Signed-off-by: Russell King (Oracle) <rmk+kernel@armlinux.org.uk>
Reviewed-by: Maxime Chevallier <maxime.chevallier@bootlin.com>
Link: https://patch.msgid.link/E1vkLmM-00000005vE1-0nop@rmk-PC.armlinux.org.uk
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Merge series from "Anton D. Stavinskii" <stavinsky@gmail.com>:
This patch series adds basic audio support for Sophgo CV1800B,
as used on boards such as the Milk-V Duo.
The series introduces the I2S controller driver,
the DAC and ADC codec drivers, corresponding DT bindings,
and DTS updates to wire the components together.
The implementation is based on vendor documentation
and testing on real hardware. This series relies on
recent fixes in the DesignWare AXI DMA support;
in particular, correct operation depends on
the DMA changes discussed at:
https://lore.kernel.org/all/20251214224601.598358-1-inochiama@gmail.com/
The current driver implementation supports a fixed audio configuration
of 48 kHz sample rate and only I2S protocol which is used in codecs.
The series has been tested on the Milk-V Duo 256M board using
the Sophgo SG2002 SoC. The implementation is expected to also work on
Milk-V Duo and Milk-V Module boards based on the SG2000 SoC,
as the audio and DMA blocks are closely related.
Known hardware limitation:
On CV1800B / SG2002, the I2S2 output pins cannot be enabled via
pinctrl alone. Due to SoC design constraints, the output path becomes
functional only after additional vendor-specific register programming.
This series makes the limitation explicit and does not attempt
to work around it implicitly via pinctrl or undocumented behavior.
Merge series from Kuninori Morimoto <kuninori.morimoto.gx@renesas.com>:
Now, all DAPM users are using function to handling it.
We can capsuling it. This patchset moves struct snd_soc_dapm_context into
soc-dapm.c, and remove un-used functions.
Link: https://lore.kernel.org/r/87zf7jrx52.wl-kuninori.morimoto.gx@renesas.com
Merge series from "Peng Fan (OSS)" <peng.fan@oss.nxp.com>:
The main goal is to convert drivers to use GPIO descriptors. While reading
the code, I think it is time to remove ak4641 and pxa2xx-ac97 driver,
more info could be found in commit log of each patch.
Then only need to convert sound/arm/pxa2xx-ac97-lib.c to use GPIO
descriptors. Not have hardware to test the pxa2xx ac97.
Merge series from Shengjiu Wang <shengjiu.wang@nxp.com>:
Enable AUDMIX on i.MX952 platform, update dt binding document and
driver.
SAI is connected to AUDMIX, and the AUDMIX can be bypassed, so
add 'fsl,sai-amix-mode' property in SAI binding document for this
case.
Make the source folders a dependency for the generated folder root so
that whenever a file is deleted from the source it will force a new
fresh copy of all the JSON files and avoid stale deleted files.
JSON_DIRS_OUTPUT_ROOT needs to be a dependency of LEGACY_CACHE_JSON so
that the root folder doesn't get cleaned after the legacy JSON is
generated. But this is a no-op with in-source builds as
JSON_DIRS_OUTPUT_ROOT is unset.
JSON_DIRS is added as a dependency of PMU_EVENTS_C which also forces a
re-build for in source builds when JSON files are deleted. This could
have also resulted in stale builds, but never a broken one.
Closes: https://lore.kernel.org/linux-next/aW5XSAo88_LBPSYI@sirena.org.uk/
Fixes: 4bb55de4ff ("perf jevents: Support copying the source json files to OUTPUT")
Reported-by: Mark Brown <broonie@kernel.org>
Signed-off-by: James Clark <james.clark@linaro.org>
Cc: Adrian Hunter <adrian.hunter@intel.com>
Cc: Alexander Shishkin <alexander.shishkin@linux.intel.com>
Cc: Ian Rogers <irogers@google.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Jiri Olsa <jolsa@kernel.org>
Cc: Mark Rutland <mark.rutland@arm.com>
Cc: Namhyung Kim <namhyung@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
Ian Rogers noticed that 678ed6b707 ("perf strlist: Don't write to
const memory") breaks the 'Remove thread map' 'perf test' entry, because
it keeps pointers to the temporary string introduced to avoid touching
the const memory.
This is because the thread_map__new_by_[pt]id_str() were the only
methods using the slist->dont_dupstr knob to keep pointers to the
original const string list, as it uses strtol to parse numbers and it
stops at the comma.
As this is the only case of dont_dupstr use, dupstr being the default,
and it gets in the way of getting rid of the last const-correctness,
remove this knob, with it:
$ perf test 37
37: Remove thread map : Ok
$
Fixes: 678ed6b707 ("perf strlist: Don't write to const memory")
Reported-by: Ian Rogers <irogers@google.com>
Tested-by: Ian Rogers <irogers@google.com>
Signed-off-by: Arnaldo Carvalho de Melo <acme@redhat.com>
This eliminates the need for `expect_punct` helper.
Reviewed-by: Tamir Duberstein <tamird@gmail.com>
Reviewed-by: Benno Lossin <lossin@kernel.org>
Signed-off-by: Gary Guo <gary@garyguo.net>
Link: https://patch.msgid.link/20260112170919.1888584-8-gary@kernel.org
Signed-off-by: Miguel Ojeda <ojeda@kernel.org>
This has no behavioural change, but is good for maintainability. With
`quote!`, we're no longer using string templates, so we don't need to
quote " and {} inside the template anymore. Further more, editors can
now highlight the code template.
This also improves the robustness as it eliminates the need for string
quoting and escaping.
Co-developed-by: Benno Lossin <lossin@kernel.org>
Signed-off-by: Benno Lossin <lossin@kernel.org>
Signed-off-by: Gary Guo <gary@garyguo.net>
Link: https://patch.msgid.link/20260112170919.1888584-6-gary@kernel.org
Signed-off-by: Miguel Ojeda <ojeda@kernel.org>
With `syn` being available in the kernel, use it to parse the complex
custom `module!` macro to replace existing helpers. Only parsing is
changed in this commit, the code generation is untouched.
This has the benefit of better error message when the macro is used
incorrectly, as it can point to a concrete span on what's going wrong.
For example, if a field is specified twice, previously it reads:
error: proc macro panicked
--> samples/rust/rust_minimal.rs:7:1
|
7 | / module! {
8 | | type: RustMinimal,
9 | | name: "rust_minimal",
10 | | author: "Rust for Linux Contributors",
11 | | description: "Rust minimal sample",
12 | | license: "GPL",
13 | | license: "GPL",
14 | | }
| |_^
|
= help: message: Duplicated key "license". Keys can only be specified once.
now it reads:
error: duplicated key "license". Keys can only be specified once.
--> samples/rust/rust_minimal.rs:13:5
|
13 | license: "GPL",
| ^^^^^^^
Reviewed-by: Tamir Duberstein <tamird@gmail.com>
Signed-off-by: Gary Guo <gary@garyguo.net>
Reviewed-by: Benno Lossin <lossin@kernel.org>
Link: https://patch.msgid.link/20260112170919.1888584-5-gary@kernel.org
Signed-off-by: Miguel Ojeda <ojeda@kernel.org>
`#[vtable]` is converted to use syn. This is more robust than the
previous heuristic-based searching of defined methods and functions.
When doing so, the trait and impl are split into two code paths as the
types are distinct when parsed by `syn`.
Reviewed-by: Tamir Duberstein <tamird@gmail.com>
Signed-off-by: Gary Guo <gary@garyguo.net>
Reviewed-by: Benno Lossin <lossin@kernel.org>
Link: https://patch.msgid.link/20260112170919.1888584-4-gary@kernel.org
Signed-off-by: Miguel Ojeda <ojeda@kernel.org>
With `quote` crate now vendored in the kernel, we can remove our custom
`quote!` macro implementation and just rely on that crate instead.
The `quote` crate uses types from the `proc-macro2` library so we also
update to use that, and perform conversion in the top-level lib.rs.
Clippy complains about unnecessary `.to_string()` as `proc-macro2`
provides additional `PartialEq` impl, so they are removed.
Reviewed-by: Tamir Duberstein <tamird@gmail.com>
Reviewed-by: Benno Lossin <lossin@kernel.org>
Signed-off-by: Gary Guo <gary@garyguo.net>
Acked-by: David Gow <davidgow@google.com> # for kunit
Link: https://patch.msgid.link/20260112170919.1888584-3-gary@kernel.org
Signed-off-by: Miguel Ojeda <ojeda@kernel.org>
The first byte of the Rx frame is a copy of the Rx status register, so
0x40 corresponds to RSR_MF (meaning the frame is multicast). Replace
0x40 with RSR_MF for clarity. (All other bits of the RSR indicate
errors. The fact that the driver ignores these errors will be fixed by
a later patch.)
The first byte of the status URB is a copy of the NSR, so 0x40
corresponds to NSR_LINKST. Replace 0x40 with NSR_LINKST for clarity.
Signed-off-by: Ethan Nelson-Moore <enelsonmoore@gmail.com>
Reviewed-by: Peter Korsgaard <peter@korsgaard.com>
Link: https://patch.msgid.link/20260124032248.26807-1-enelsonmoore@gmail.com
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
Now that there are no users of the low-level SHA-1 interface, remove it.
Specifically:
- Remove SHA1_DIGEST_WORDS (no longer used)
- Remove sha1_init_raw() (no longer used)
- Rename sha1_transform() to sha1_block_generic() and make it static
- Move SHA1_WORKSPACE_WORDS into lib/crypto/sha1.c
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Eric Biggers <ebiggers@kernel.org>
Link: https://patch.msgid.link/20260123051656.396371-3-ebiggers@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
There's now a proper SHA-1 API that follows the usual conventions for
hash function APIs: sha1_init(), sha1_update(), sha1_final(), sha1().
The only remaining user of the older low-level SHA-1 API,
sha1_init_raw() and sha1_transform(), is ipv6_generate_stable_address().
I'd like to remove this older API, which is too low-level.
Unfortunately, ipv6_generate_stable_address() does in fact skip the
SHA-1 finalization for some reason. So the values it computes are not
standard SHA-1 values, and it sort of does want the low-level API.
Still, it's still possible to use the higher-level functions sha1_init()
and sha1_update() to get the same result, provided that the resulting
state is used directly, skipping sha1_final().
So, let's do that instead. This will allow removing the low-level API.
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Signed-off-by: Eric Biggers <ebiggers@kernel.org>
Acked-by: David Ahern <dsahern@kernel.org>
Link: https://patch.msgid.link/20260123051656.396371-2-ebiggers@kernel.org
Signed-off-by: Jakub Kicinski <kuba@kernel.org>
The power state check in amdgpu_dpm_set_powergating_by_smu() is done
before acquiring the pm mutex, leading to a race condition where:
1. Thread A checks state and thinks no change is needed
2. Thread B acquires mutex and modifies the state
3. Thread A returns without updating state, causing inconsistency
Fix this by moving the mutex lock before the power state check,
ensuring atomicity of the state check and modification.
Fixes: 6ee27ee27b ("drm/amd/pm: avoid duplicate powergate/ungate setting")
Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 7a3fbdfd19)
On APUs such as Raven and Renoir (GC 9.1.0, 9.2.2, 9.3.0), the ih1 and
ih2 interrupt ring buffers are not initialized. This is by design, as
these secondary IH rings are only available on discrete GPUs. See
vega10_ih_sw_init() which explicitly skips ih1/ih2 initialization when
AMD_IS_APU is set.
However, amdgpu_gmc_filter_faults_remove() unconditionally uses ih1 to
get the timestamp of the last interrupt entry. When retry faults are
enabled on APUs (noretry=0), this function is called from the SVM page
fault recovery path, resulting in a NULL pointer dereference when
amdgpu_ih_decode_iv_ts_helper() attempts to access ih->ring[].
The crash manifests as:
BUG: kernel NULL pointer dereference, address: 0000000000000004
RIP: 0010:amdgpu_ih_decode_iv_ts_helper+0x22/0x40 [amdgpu]
Call Trace:
amdgpu_gmc_filter_faults_remove+0x60/0x130 [amdgpu]
svm_range_restore_pages+0xae5/0x11c0 [amdgpu]
amdgpu_vm_handle_fault+0xc8/0x340 [amdgpu]
gmc_v9_0_process_interrupt+0x191/0x220 [amdgpu]
amdgpu_irq_dispatch+0xed/0x2c0 [amdgpu]
amdgpu_ih_process+0x84/0x100 [amdgpu]
This issue was exposed by commit 1446226d32 ("drm/amdgpu: Remove GC HW
IP 9.3.0 from noretry=1") which changed the default for Renoir APU from
noretry=1 to noretry=0, enabling retry fault handling and thus
exercising the buggy code path.
Fix this by adding a check for ih1.ring_size before attempting to use
it. Also restore the soft_ih support from commit dd29944165 ("drm/amdgpu:
Rework retry fault removal"). This is needed if the hardware doesn't
support secondary HW IH rings.
v2: additional updates (Alex)
Closes: https://gitlab.freedesktop.org/drm/amd/-/issues/3814
Fixes: dd29944165 ("drm/amdgpu: Rework retry fault removal")
Reviewed-by: Timur Kristóf <timur.kristof@gmail.com>
Reviewed-by: Philip Yang <Philip.Yang@amd.com>
Signed-off-by: Jon Doron <jond@wiz.io>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 6ce8d536c8)
Cc: stable@vger.kernel.org
v1:
resolve the issue where some freq frequencies cannot be set correctly
due to insufficient floating-point precision.
v2:
patch this convert on 'max' value only.
Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 53868dd877)
Cc: stable@vger.kernel.org
v1:
resolve the issue where some freq frequencies cannot be set correctly
due to insufficient floating-point precision.
v2:
patch this convert on 'max' value only.
Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Alex Deucher <alexander.deucher@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
(cherry picked from commit 6194f60c70)
Cc: stable@vger.kernel.org
Added:
- '&'static mut MaybeUninit<T>' now implements 'InPlaceWrite'. This
enables users to use external allocation mechanisms such as
'static_cell'.
- Gary Guo as a Maintainer.
Changed:
- Rewrote all proc-macros ('[pin_]init!', '#[pin_data]',
'#[pinned_drop]', 'derive([Maybe]Zeroable)'), using 'syn' with better
diagnostics.
- 'derive([Maybe]Zeroable)' now support tuple structs.
- '[pin_]init!' now supports attributes on fields (such as
'#[cfg(...)]').
- Add a '#[default_error(<type>)]' attribute to '[pin_]init!' to
override the default error (when no '? Error' is specified).
- Support packed struct in '[pin_]init!' with
'#[disable_initialized_field_access]'.
Removed:
- 'try_[pin_]init!' have been removed in favor of merging their feature
with '[pin_]init!'. The kernel's own 'try_[pin_]init!' macros have
been updated to use the 'default_error' attribute.
Fixed:
- Corrected 'T: Sized' bounds to 'T: ?Sized' in the generated
'PinnedDrop' check by '#[pin_data]'.
-----BEGIN PGP SIGNATURE-----
iIgEABYKADAWIQQjEG/HT3UeYLJybLTomd21rZaLygUCaXHz9BIcbG9zc2luQGtl
cm5lbC5vcmcACgkQ6Jndta2Wi8q1AQD+IAf84ueKJ8vEfiSRP0IgteZakHxFHCyZ
Uqsn0MQRUTgA/238IG65zYkrT8jaT3FY+LSoaNRY4rNS8/l9bxFWqqwA
=dBj9
-----END PGP SIGNATURE-----
Merge tag 'pin-init-v7.0' of https://github.com/Rust-for-Linux/linux into rust-next
Pull pin-init updates from Benno Lossin:
"Added:
- Implement 'InPlaceWrite' for '&'static mut MaybeUninit<T>'. This
enables users to use external allocation mechanisms such as
'static_cell'.
- Add Gary Guo as a Maintainer.
Changed:
- Rewrote all proc-macros ('[pin_]init!', '#[pin_data]',
'#[pinned_drop]', 'derive([Maybe]Zeroable)'), using 'syn' with
better diagnostics.
- Support tuple structs in 'derive([Maybe]Zeroable)'.
- Support attributes on fields in '[pin_]init!' (such as
'#[cfg(...)]').
- Add a '#[default_error(<type>)]' attribute to '[pin_]init!' to
override the default error (when no '? Error' is specified).
- Support packed structs in '[pin_]init!' with
'#[disable_initialized_field_access]'.
Removed:
- Remove 'try_[pin_]init!' in favor of merging their feature
with '[pin_]init!'. Update the kernel's own 'try_[pin_]init!'
macros to use the 'default_error' attribute.
Fixed:
- Correct 'T: Sized' bounds to 'T: ?Sized' in the generated
'PinnedDrop' check by '#[pin_data]'."
* tag 'pin-init-v7.0' of https://github.com/Rust-for-Linux/linux:
rust: pin-init: Implement `InPlaceWrite<T>` for `&'static mut MaybeUninit<T>`
MAINTAINERS: add Gary Guo to pin-init
rust: pin-init: internal: init: simplify Zeroable safety check
rust: pin-init: internal: init: add escape hatch for referencing initialized fields
rust: pin-init: internal: init: add support for attributes on initializer fields
rust: init: use `#[default_error(err)]` for the initializer macros
rust: pin-init: add `#[default_error(<type>)]` attribute to initializer macros
rust: pin-init: rewrite the initializer macros using `syn`
rust: pin-init: add `?Sized` bounds to traits in `#[pin_data]` macro
rust: pin-init: rewrite `#[pin_data]` using `syn`
rust: pin-init: rewrite the `#[pinned_drop]` attribute macro using `syn`
rust: pin-init: rewrite `derive(Zeroable)` and `derive(MaybeZeroable)` using `syn`
rust: pin-init: internal: add utility API for syn error handling
rust: pin-init: add `syn` dependency and remove `proc-macro[2]` and `quote` workarounds
rust: pin-init: allow the crate to refer to itself as `pin-init` in doc tests
rust: pin-init: remove `try_` versions of the initializer macros
Add additional capabilities reporting.
Signed-off-by: Jonathan Kim <jonathan.kim@amd.com>
Reviewed-by: James Zhu <james.zhu@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Some older builds weren't sending RMA CPERs when the bad page threshold
was exceeded. Newer builds have resolved this, but there could be
systems out there with bad page numbers higher than the threshold, that
haven't sent out an RMA CPER. To be thorough and safe, send an RMA CPER
when we load the table, if the threshold is met or exceeded, instead of
waiting for the next UE to trigger the CPER.
Signed-off-by: Kent Russell <kent.russell@amd.com>
Reviewed-by: Tao Zhou <tao.zhou1@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Fix the vcn reset sequence in vcn_v4_0_3_ring_reset() to restore
JPEG power state and unlock the JPEG powergating mutex before
running the JPEG ring post-reset helper.
Fixes: d25c67fd9d ("drm/amdgpu/vcn4.0.3: rework reset handling")
Reviewed-by: Lijo Lazar <lijo.lazar@amd.com>
Signed-off-by: Jesse Zhang <jesse.zhang@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
The power state check in amdgpu_dpm_set_powergating_by_smu() is done
before acquiring the pm mutex, leading to a race condition where:
1. Thread A checks state and thinks no change is needed
2. Thread B acquires mutex and modifies the state
3. Thread A returns without updating state, causing inconsistency
Fix this by moving the mutex lock before the power state check,
ensuring atomicity of the state check and modification.
Fixes: 6ee27ee27b ("drm/amd/pm: avoid duplicate powergate/ungate setting")
Signed-off-by: Yang Wang <kevinyang.wang@amd.com>
Reviewed-by: Kenneth Feng <kenneth.feng@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
* Fw release 0.1.44.0
* Fixes for corruption on platforms older than DCN4x.
* Bug fixes related to USB4 link training
* Fixes related to FP guard
* Debug helpers and other stability fixes.
* Some refactors to improve code quality
Signed-off-by: Taimur Hassan <Syed.Hassan@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Reviewed-by: Alex Hung <alex.hung@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[why]
Direct HUBBUB register access in the hwseq layer was creating register conflicts.
[how]
Migrated HUBBUB registers from hwseq to the hubbub component.
Reviewed-by: Martin Leung <martin.leung@amd.com>
Signed-off-by: Bhuvanachandra Pinninti <bpinnint@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[WHY]
Enable dynamic ODM testing without needing a valid dispclk table
[HOW]
Create a debug flag to specify an override value for min dispclk
Reviewed-by: Dmytro Laktyushkin <dmytro.laktyushkin@amd.com>
Signed-off-by: Michael Strauss <michael.strauss@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why]
When usb4 link training fails, the dpia sym clock will be disabled and SYMCLK
source should be changed back to phy clock. In enable_streams, it is
assumed that link training succeeded and will switch from refclk to
phy clock. But phy clk here might not be on. Dig reg access timeout
will occur.
[How]
When enable_stream is hit, check if link training failed for usb4.
If it did, fall back to the ref clock to avoid reg access timeout.
Reviewed-by: Wenjing Liu <wenjing.liu@amd.com>
Signed-off-by: Zhongwei <Zhongwei.Zhang@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[Why & How]
For dcn2x_fast_validate_bw(), not only populate_dml_pipes needs FP guard
but also dml_get_voltage_level().
Remove unnecessary DC_FP_START/DC_FP_END guard in dcn20_fast_validate_bw
and dcn21_fast_validate_bw. FP guard is already there before calling
dcn2x_validate_bandwidth_fp().
Reviewed-by: ChiaHsuan (Tom) Chung <chiahsuan.chung@amd.com>
Signed-off-by: Wayne Lin <Wayne.Lin@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
[why]
stereo_sync pin is removed, but we still support display stereo
Reviewed-by: Ovidiu (Ovi) Bunea <ovidiu.bunea@amd.com>
Signed-off-by: Charlene Liu <Charlene.Liu@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
Existing version check that limits the sequence to clear update flags
should be performed for all asics. Exclude DCE asics for now.
Reviewed-by: Sun peng (Leo) Li <sunpeng.li@amd.com>
Signed-off-by: Aurabindo Pillai <aurabindo.pillai@amd.com>
Tested-by: Dan Wheeler <daniel.wheeler@amd.com>
Signed-off-by: Alex Deucher <alexander.deucher@amd.com>
