In Linux v6.6, the crypto_engine_ctx structure was removed and the
crypto_engine_ops was moved from this structure to the crypto_alg
structure. The Tegra Security Engine driver was updated accordingly to
populate the crypto_engine_ops in the appropriate structure depending
on whether the crypto_engine_ctx structure is present or not.
Currently conftest is using the presence of the crypto_engine_ctx
structure to decide where to populate the crypto_engine_ops. While
this works, it is possible that a kernel older than v6.6 also includes
a backport of commit e5e7eb023f24 ("crypto: engine - Move
crypto_engine_ops from request into crypto_alg") that moves the ops
to the crypto_alg structure. Although backporting this commit alone
should not cause any problems, it is better to detect if this commit
is present rather than relying on the presence of crypto_engine_ctx
structure.
To detect the presence of commit e5e7eb023f24 we can simply detect if
one of the functions added and exported by this commit is present.
For example, the function crypto_engine_register_aead() was added and
exported by this commit. Therefore, update conftest to detect if
commit e5e7eb023f24 is present and then update the Tegra Security
Engine accordingly.
Bug 5564561
Change-Id: I3f7688e6e954a158094a9a2fdf73354f765ed680
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3499816
(cherry picked from commit c1631d32f8d80cd31f54e7297c542f308a281d25)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3508186
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Brad Griffis <bgriffis@nvidia.com>
In Linux v6.3, commit 03c835f498b5 ("i2c: Switch .probe() to not take an
id parameter") removed the ID parameter from the I2C probe function. The
max96712 does not build for Linux kernels prior to Linux v6.3 because it
assumes that the ID parameter has been removed. There is an existing
conftest that checks to see if the probe function requires and ID
parameter and so update the max967xx drivers accordingly to fix the build
for kernels prior to Linux v6.3.
Note we should avoid using explicit kernel version checks in out-of-tree
drivers because this does not work for 3rd party kernels. Therefore,
remove the kernel version.h because this is not used or needed.
Bug 5277201
Change-Id: Idbc569d782a952ee9ea0a91aace54df20f50701d
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3506342
Tested-by: Shawn Bai <xueyuanb@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Shawn Bai <xueyuanb@nvidia.com>
- snd_soc_dapm_dai_get_connected_widgets returns
number of valid paths which is being returned
as an error code to the mixer control put callbacks
of codec master mode and frame mode controls leading
to them not set properly.
- If the master mode mixer control is set before the
frame mode control, this issue will not be observed
as the fmt value is still updated in hardware. User
space though, will show a wrong value.
- Internal I2S loopback failed in Android sanity
as the frame mode control is set before
the master mode.
- Modify the return value to 0 if it is >0
to fix this issue.
Bug 5705385
Change-Id: I6662057503b9d5be6b71dfdb219334267ea9b9bc
Signed-off-by: Aditya Bavanari <abavanari@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3503844
Reviewed-by: Sameer Pujar <spujar@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Tested-by: Sheetal . <sheetal@nvidia.com>
Reviewed-by: Mohan kumar <mkumard@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Sheetal . <sheetal@nvidia.com>
Fix a memory leak in the NVIDIA Tegra ASoC machine driver where
allocated DAPM widget lists were never freed. The function
dpcm_runtime_set_dai_fmt() calls snd_soc_dapm_dai_get_connected_widgets()
which allocates memory for a widget list, but was returning without
calling snd_soc_dapm_dai_free_widgets() to free this memory.
The leak was detected by kmemleak:
unreferenced object 0xffff00008fc7e300 (size 192):
comm "alsactl", pid 812, jiffies 4294900395
backtrace:
kmemleak_alloc+0xc0/0xe8
__kmalloc+0x2e0/0x4c8
snd_soc_dapm_dai_get_connected_widgets+0x108/0x2a0
Fix this by using snd_soc_dapm_dai_free_widgets() on
success or failure to ensure the widget list is always freed
before returning.
Bug 5667057
Change-Id: Ied69968884fc0646a3041e54427736ff1fc232ac
Signed-off-by: Sheetal <sheetal@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3500426
(cherry picked from commit 14b3fc2296967968d5972054bff38cf1bf4b518e)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3502770
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Sameer Pujar <spujar@nvidia.com>
Reviewed-by: Mohan kumar <mkumard@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
When loading the WCH driver on IGX platforms the various information is
printed to the kernel log using printk. However, because no level is
specified in printk statements, the prints appear as warnings even
though they are informational. The WCH driver implements some
definitions to enable/disable prints and so wrap these informational
prints with WCH_DBG so that they are disabled by default.
Bug 5699474
Change-Id: Ic1426ceab94cd457a1ebdfb57a55ff305ad58ff9
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3500555
(cherry picked from commit c3a03860ada48533d2f81bf33ada02c44d079d9f)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3502173
Reviewed-by: svcacv <svcacv@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Brad Griffis <bgriffis@nvidia.com>
This commit implements T264 ISP floorsweeping functionality in the Linux
kernel driver, extending the existing RCE firmware implementation to
provide transparent ISP unit redirection at the kernel level.
Changes:
- Add isp_capture_query_availability_mask_probe() function to query ISP
availability mask from RCE firmware during driver probe via IVC
- Add isp_capture_find_first_available() helper function to find the
first available ISP unit from a bitmask for HW assignment
- Integrate floorsweeping logic into isp_get_nvhost_device() to handle
SW-to-HW ISP unit mapping with thread-safe assignment
- Add isp_capture_get_capabilities() function to expose ISP availability
to userspace via new IOCTL (ISP_CAPTURE_GET_CAPABILITIES)
- Add probe-time caching of ISP availability mask to avoid repeated IVC
queries during runtime
- Add comprehensive error handling and safety mechanisms for unknown
configurations
Implementation details:
- Uses CAPTURE_ISP_FUSE_QUERY_REQ/RESP IVC messages for querying RCE
- Maintains SW-to-HW ISP unit mapping array protected by mutex
- The SW-to-HW mappings persist for the driver's lifetime, given that
multiple channels make use of the same SW unit ID's and we do not
have tracking of all open channels during runtime
- Integrates with existing ISP setup flow before buffer allocation
- Maintains backward compatibility with non-T264 hardware
- Conservative approach: blocks requests when no ISP units are available
Floorsweeping logic (handled by RCE firmware, exposed as bitmask):
- Mask bit N set = ISP N available (not fused off)
- Mask bit N clear = ISP N fused off (unavailable)
- Examples:
- 0x3 (0b11): Both ISP0/ISP1 available (no floorsweeping)
- 0x1 (0b01): Only ISP0 available → all requests map to ISP0
- 0x2 (0b10): Only ISP1 available → all requests map to ISP1
- 0x0: No ISP units available (block all requests)
Benefits:
- Transparent operation for applications using fusacapture library
- Automatic ISP unit redirection without API changes
- Improved hardware yield by supporting single-ISP configurations
- Robust error handling and safety mechanisms
- Minimal performance overhead with probe-time caching
The implementation complements the existing RCE firmware floorsweeping
support and provides a complete end-to-end solution for T264 ISP
floorsweeping across all software layers.
Bug 5640706
Change-Id: If85ee3178e857394300479dd42e636f5f5d3bd23
Signed-off-by: fraunak <fraunak@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3491180
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Evgeny Kornev <ekornev@nvidia.com>
Reviewed-by: Kalle Jokiniemi <kjokiniemi@nvidia.com>
Reviewed-by: Semi Malinen <smalinen@nvidia.com>
Tested-by: Akihiro Mizusawa <amizusawa@nvidia.com>
Add gpu-static-pg kernel driver to support the
BPMP GPC/TPC/FBP setting via sysfs communication
and module parameter, which allow the nvpmodel
system service to set the mask via the interfaces
above.
For nvpmodel service usecase, this driver should
be probed and the mask should be applied before
the GPU driver is probed.
Bug 5516617
Signed-off-by: Shao-Chun Kao <shaochunk@nvidia.com>
Change-Id: Id16e374c21acccd08c0c33c2df1aa909fcba5658
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3458914
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Yi-Wei Wang <yiweiw@nvidia.com>
Reviewed-by: Bibek Basu <bbasu@nvidia.com>
WLAN_AKM_SUITE_FT_OVER_SAE(0x000FAC09) is not processed in
rtw_cfg80211_set_key_mgt, which makes rtw_ft_validate_akm_type not set
RTW_FT_PEER_EN to ft_flags, and thus MDIE is not built in
rtw_ft_build_assoc_req_ies
Fix:
- Add WLAN_AKM_SUITE_FT_OVER_SAE into rtw_cfg80211_set_key_mgt check condition
- Add 9 into rtw_ft_valid_akm check condition
- Update driver version number to 277-9-6
Bug 5580277
Change-Id: I7cc987a9141dc65577dc20e804e92def10e9412e
Signed-off-by: Shaofu <shaofu@realtek.com>
Signed-off-by: Narayana Reddy P <narayanr@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3485741
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
Reviewed-by: Shobek Attupurath <sattupurath@nvidia.com>
----------------------------------
rtl8852ce release notes for Nvidia
----------------------------------
v1.19.16_nv-277-9-0-g3c314be22.20251020_Certified_Module
- same as v1.19.16_nv-277-9-g3c314be22.20250903_Certified_Module_beta
- just remove "beta" string for formal release
==================================================================
v1.19.16_nv-277-9-1-gc6d0f2e20.20251015_Certified_Module_beta
* If ROAM CMD (wpa_cli roam), do scan if bssid not found even in busy traffic
Tested:
- simple connection test: PASS
- wpa_cli roam command is work and show prev_bssid in log: PASS
==================================================================
v1.19.16_nv-277-14-ga7d65c4a9.20251002_Certified_Module_beta
* fix the deauth reason 7 caused by roaming racing
* downgrade the fwstate debug level
==================================================================
v1.19.16_nv-277-12-g1a8402802.20250926_Certified_Module_beta
* add fwstate debug log
==================================================================
v1.19.16_nv-277-11-g3b6c1d36e.20250925_Certified_Module_beta
* default enable ACS and RSSI debug
* add reauth and reassoc timeout config via insmod or /proc/
==================================================================
Bug 5504994
Change-Id: I3ecd41fdb28de12bf6e37d9dfaca5c7f40e2fe1d
Signed-off-by: Narayana Reddy P <narayanr@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3473467
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Shobek Attupurath <sattupurath@nvidia.com>
Reviewed-by: Bibek Basu <bbasu@nvidia.com>
During system shutdown, a kernel crash could occur because the r8126
driver's tally counter update routine (`rtl8126_dump_tally_counter`)
was still accessing PCIe address space after the device was already
stopped or powered down.
To prevent this, introduce a new driver flag `R8126_FLAG_SHUTDOWN`
that is set in `rtl8126_shutdown()`. This flag is checked
now before accessing tally counters, ensuring no further
crash during shutdown.
Bug 5620576
Change-Id: I2bb74fc52e712198bfe15b109a2136dc8a2affcd
Signed-off-by: Revanth Kumar Uppala <ruppala@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3482325
Reviewed-by: Ashutosh Jha <ajha@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Ninad Malwade <nmalwade@nvidia.com>
Tested-by: Ninad Malwade <nmalwade@nvidia.com>
Reviewed-by: Shobek Attupurath <sattupurath@nvidia.com>
In the previous release for T264 platforms,the restart lane bringup
logic was controlled through the device tree flag
'nvidia,pcs-rx-eq-sw-ovrd'.
In incremental releases, the flag has been renamed
'nvidia,force-restart-lane-bringup'.
To maintain backward compatibility with older device trees,
the driver now checks for both flags.If either flag is present,
restart lane bringup is executed.
The legacy flag 'nvidia,pcs-rx-eq-sw-ovrd' is deprecated.
Bug 5017313
Change-Id: I24040f508ef776e29a0bb0c1f07d4a74fa4cc8cc
Signed-off-by: Revanth Kumar Uppala <ruppala@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3471233
Reviewed-by: Srinivas Ramachandran <srinivasra@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
When BWMGR is unavailable, BWMGR could be either unsupported and
disabled in BPMP firmware.
For unsupported state, MRQ_BWMGR_INT is not available to use.
For disabled state, MRQ_BWMGR_INT is available but sending any
MRQ_BWMGR_INT request will just get -BPMP_NODEV error code.
Add error handling logic to make min_freq and max_freq devfreq sysfs
nodes as read-only but still expose devfreq nodes, such as
available_frequencies, min_freq, and max_freq, to allow user space
apps to read information from.
Bug 5555075
Change-Id: Id7c2695765d3dfee148a0d1bc1f7b0552fe4b343
Signed-off-by: Johnny Liu <johnliu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3467128
Reviewed-by: svcacv <svcacv@nvidia.com>
Reviewed-by: Rajkumar Kasirajan <rkasirajan@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Replace struct hack pattern with flexible array members in msgq_t
and msgq_message_t to resolve UBSAN warnings.
The message queue implementation was using the old "struct hack"
pattern with single-element arrays (int32_t queue[1] and
int32_t payload[1]) to create variable-length structures. While
functionally correct, this triggers UBSAN array-index-out-of-bounds
errors when accessing elements beyond index 0, even though the
memory is properly allocated.
UBSAN errors observed:
- msgq.c:64: index 2045 out of range for type 'int32_t [1]'
- msgq.c:69: index 153 out of range for type 'int32_t [1]'
- msgq.c:124: index 53 out of range for type 'int32_t [1]'
- msgq.c:148: index 53 out of range for type 'int32_t [1]'
- msgq.c:149: index 2045 out of range for type 'int32_t [1]'
Changes:
1. Convert queue[1] to queue[] in msgq_t structure
2. Convert payload[1] to payload[] in msgq_message_t structure
3. Update MSGQ_HEADER_SIZE and MSGQ_MESSAGE_HEADER_SIZE macros
to use sizeof() directly, as flexible array members have zero
size and cannot be used with sizeof()
The flexible array member (FAM) approach is:
- C99 standard compliant
- Linux kernel best practice for variable-length structures
- Binary compatible with the previous implementation
- Eliminates UBSAN false positives without functional changes
Bug 4831393
Change-Id: I243d4a1b1f091bf17cfc10337e75dbd1b878042f
Signed-off-by: Asha Talambedu <atalambedu@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3464624
Reviewed-by: Mohan kumar <mkumard@nvidia.com>
Reviewed-by: Viswanath L <viswanathl@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
The MTTCAN controller uses different core clock frequencies depending on
the SoC variant: 40MHz for T264 and 50MHz for T194/T234. The 8Mbps data
bitrate needs to be an exact divisor of the clock frequency for proper
operation.
For 40MHz clock: 8Mbps = 8000000 bps (40MHz / 5)
For 50MHz clock: 8Mbps = 8333333 bps (50MHz / 6)
The previous single MTTCAN_SPEED_8MBPS definition (8333333) only worked
correctly for 50MHz clocks and would not divide evenly for 40MHz clocks.
Split MTTCAN_SPEED_8MBPS into two separate defines:
- MTTCAN_SPEED_8MBPS_40MHZ: 8000000 for exact 8Mbps on 40MHz clock
- MTTCAN_SPEED_8MBPS_50MHZ: 8333333 for closest 8Mbps on 50MHz clock
Both values use the same production settings (prod_c_can_8m) and will be
selected based on the chip's configured clock frequency. The 5Mbps rate
remains unchanged as it divides evenly for both clock frequencies.
Bug 5497458
Change-Id: I17190ae6821f89d33e76114f20b91dbd65ba25f2
Signed-off-by: Shubhi Garg <shgarg@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3469194
Reviewed-by: Bibek Basu <bbasu@nvidia.com>
Tested-by: Kevin Fu <chunhuaif@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: svcacv <svcacv@nvidia.com>
We cannot use enum definitions in '#if defined()' statements because the
'#if defined()' is a preprocessor macro that is evaluated prior to
compilation. Fix this by adding a test to conftest to detect if
CMD_PCIE_RP_CONTROLLER_OFF is defined.
Bug 5551652
Change-Id: I71826a58e428f1a335e8e0d895f25f3cea46a65e
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3467641
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Prakhar Srivastava (SW-TEGRA) <prasrivastav@nvidia.com>
Upstream commit a55893133830 ("gpiolib: Remove unused
devm_gpio_request()") removed the devm_gpio_request() function and this
breaks the Tegra USS IO Proxy driver. The devm_gpiod_get() function has
been supported in the Linux kernel since v4.3 and now that the legacy
GPIO functions are being removed, migrate the Tegra USS IO Proxy driver
over to using GPIO descriptors.
Bug 4387902
Bug 5420210
Change-Id: I0c62977dc4ec829e14f105babdd08bbd64113fa2
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3423827
(cherry picked from commit e4572a5ff434f778178b1dfb39b4c1cf7923db6d)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3462463
Reviewed-by: Brad Griffis <bgriffis@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Currently the wakeup source is configured if only the 'ext_wake' is
valid. However, the bluedroid_pm_timer and proc interface that are
configured when 'ext_wake' is valid, also assumes that 'host_wake' is
also valid. Therefore, update the code to only configure the
bluedroid_pm_timer and proc interface if both of these are valid.
Bug 4958861
Change-Id: I16ef55359c36acc9dc50464f43816ae2ef3c3123
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3397252
(cherry picked from commit b1e18d66aa0ab6d491b88caac5fa9abddb69d0e4)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3462461
Reviewed-by: Brad Griffis <bgriffis@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
The structure returned from the function wakeup_source_register() is not
currently checked to see that a valid structure is returned. If
wakeup_source_register fails to register the wake-up source, then this
function will return a NULL pointer and this will lead to a crash when
defereferencing this pointer. Therefore, check that a valid structure is
returned from wakeup_source_register() and if not then return an error
from the probe function.
JIRA LINQPJ14-60
Bug 4958861
Change-Id: I03356700ab03d6c25080b3806ea8b7da3983a302
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3394894
(cherry picked from commit 14286052fb9dd496b243221fa1066f828451b294)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3405519
Reviewed-by: Brad Griffis <bgriffis@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
The function devm_gpiod_get_optional() will return an error code encoded
as a pointer type if the GPIO requested is not found. Therefore, we need
to use the IS_ERR() macro to determine if the GPIO is valid, otherwise
we could incorrectly attempt to dereference an invalid pointer. This bug
was introduced when migrating the bluedroid driver to use the gpiod
functions.
Bug 4387902
Bug 4958861
Change-Id: Ib9e7494c92226454d93506c2c0d4c80bd6a7493c
Signed-off-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3397231
(cherry picked from commit b461fa1a87f9beb5faa1dc917956603430b12df7)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3405518
Reviewed-by: Brad Griffis <bgriffis@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
There are 2 timer defined. One statically and one in the
bluedroid_pm_data struct. Both timers call the same function on expiration. In the case of the statically defined timer this is a problem because it assumes that the timer is part of the bluedroid_pm_data and so calling timer_container_of() or from_timer() results in an invalid pointer and hence kernel panic. Fix this by
removing the statically defined timer.
Bug 4958861
Change-Id: I08f9dc3a032f84ca3350fbec5fb97062da8d6795
Signed-off-by: Bruce Xu <brucex@nvidia.com>
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3400991
(cherry picked from commit 61490c6a2a65e24068cd33a92d8730260ad83f95)
Reviewed-on: https://git-master.nvidia.com/r/c/linux-nv-oot/+/3405517
Tested-by: Jon Hunter <jonathanh@nvidia.com>
Reviewed-by: Jon Hunter <jonathanh@nvidia.com>
GVS: buildbot_gerritrpt <buildbot_gerritrpt@nvidia.com>
Reviewed-by: Brad Griffis <bgriffis@nvidia.com>
