This bug was fixed in the package linux-nvidia-6.14 - 6.14.0-1005.5
---------------
linux-nvidia-6.14 (6.14.0-1005.5) noble; urgency=medium
* noble/linux-nvidia-6.14: 6.14.0-1005.5 -proposed tracker (LP:
#2112218)
* Packaging resync (LP: #1786013)
- [Packaging] update annotations scripts
- [Packaging] update update.conf
* mt7925: Support country-specific regulatory requirements (LP: #2112155)
- wifi: mt76: mt7925: load the appropriate CLC data based on hardware type
- wifi: mt76: mt7925: add EHT control support based on the CLC data
- wifi: mt76: mt7925: update the channel usage when the regd domain changed
- wifi: mt76: mt7925: remove unused acpi function for clc
- wifi: mt76: mt792x: extend MTCL of APCI to version3 for EHT control
- wifi: mt76: mt7925: add MTCL support to enhance the regulatory compliance
- NVIDIA: SAUCE: MEDIATEK: wifi: mt76: mt7925: add reg hint support
- NVIDIA: SAUCE: MEDIATEK: wifi: mt76: mt7925: update the regd by country
code
of 11d changed
- wifi: mt76: mt7925: Fix logical vs bitwise typo
* Backport: TPM Service Command Response Buffer Interface Over FF-A
(LP: #2111511)
- tpm_crb: ffa_tpm: Implement driver compliant to CRB over FF-A
- tpm_crb: Clean-up and refactor check for idle support
- ACPICA: Add start method for ARM FF-A
- tpm_crb: Add support for the ARM FF-A start method
- Documentation: tpm: Add documentation for the CRB FF-A interface
- [Config] nvidia-6.14: Update annotations to enable TPM over FFA
* Backport: ALSA: hda - Add new driver for HDA controllers listed via ACPI
(LP: #2111447)
- ALSA: hda - Add new driver for HDA controllers listed via ACPI
- ALSA: hda: acpi: Use SYSTEM_SLEEP_PM_OPS()
- ALSA: hda: acpi: Make driver's match data const static
- NVIDIA: SAUCE: [Config] nvidia-6.14 CONFIG_SND_HDA_ACPI=m on arm64
* UBSAN: shift-out-of-bounds arm-smmu-v3.c (LP: #2110750)
- iommu/arm-smmu-v3: Fix pgsize_bit for sva domains
* Backport wifi driver changes for mt7925 (LP: #2109984)
- wifi: mt76: mt7925: fix fails to enter low power mode in suspend state
- wifi: mt76: mt7925: ensure wow pattern command align fw format
- wifi: mt76: mt7925: fix country count limitation for CLC
- wifi: mt76: Add check for devm_kstrdup()
- wifi: mt76: mt7925: Simplify HIF suspend handling to avoid suspend fail
- wifi: mt76: mt7925: fix the wrong link_idx when a p2p_device is present
* Pull request to enable GPU passthrough for CUDA (LP: #2095028)
- [Config] nvidia-6.14: Drop CONFIG_TEGRA241_CMDQV from annotations
* Pull request: Spark needs a different BT device ID (LP: #2109227)
- NVIDIA: SAUCE: Bluetooth: btusb: Add new VID/PID 13d3/3630
linux-nvidia-6.14 (6.14.0-1004.4) noble; urgency=medium
* noble/linux-nvidia-6.14: 6.14.0-1004.4 -proposed tracker (LP:
#2107819)
* Add Realtek r8127 ethernet driver (LP: #2109730)
- NVIDIA: SAUCE: r8127: Add Realtek r8127 ethernet driver
- NVIDIA: SAUCE: r8127: Remove Realtek r8127 non required files
- NVIDIA: SAUCE: r8127: Moved files from r8127/src to r8127 folder
- NVIDIA: SAUCE: Add r8127 in kernel build
- [Config] nvidia-6.11: Update annotations to enable realtek R8127 module
* Seeing warning with i2c_tegra (LP: #2109750)
- i2c: tegra: check msg length in SMBUS block read
* nvidia: take out the sbsa_gwdt of the blacklist (LP: #2109635)
- NVIDIA: [Packaging] Take out the sbsa_gwdt from the blacklist
* Pull request: Pull in the latest upstream ARM FFA patches (LP: #2109019)
- firmware: arm_ffa: Replace SCMI by FF-A in the macro
- firmware: arm_ffa: Replace UUID buffer to standard UUID format
- firmware: arm_ffa: Align sync_send_receive{,2} function prototypes
- firmware: arm_ffa: Fix big-endian support in __ffa_partition_info_get()
- firmware: arm_ffa: Fix big-endian support in
__ffa_partition_info_regs_get()
- firmware: arm_ffa: Refactor addition of partition information into XArray
- firmware: arm_ffa: Handle the presence of host partition in the partition
info
- firmware: arm_ffa: Unregister the FF-A devices when cleaning up the
partitions
- firmware: arm_ffa: Add support for passing UUID in FFA_MSG_SEND2
- firmware: arm_ffa: Upgrade FF-A version to v1.2 in the driver
- firmware: arm_ffa: Reject higher major version as incompatible
- firmware: arm_ffa: Remove unnecessary declaration of
ffa_partitions_cleanup()
- firmware: arm_ffa: Refactoring to prepare for framework notification
support
- firmware: arm_ffa: Stash ffa_device instead of notify_type in
notifier_cb_info
- firmware: arm_ffa: Add support for {un,}registration of framework
notifications
- firmware: arm_ffa: Add support for handling framework notifications
- firmware: arm_ffa: Allow multiple UUIDs per partition to register SRI
callback
- firmware: arm_ffa: Handle ffa_notification_get correctly at virtual FF-A
instance
- firmware: arm_ffa: Explicitly cast return value from FFA_VERSION before
comparison
- firmware: arm_ffa: Explicitly cast return value from NOTIFICATION_INFO_GET
- firmware: arm_ffa: Skip the first/partition ID when parsing vCPU list
- firmware: arm_ffa: Set dma_mask for ffa devices
* Packaging resync (LP: #1786013)
- [Packaging] debian.nvidia-6.14/dkms-versions -- update from
kernel-versions
(adhoc/d2025.04.04)
- [Packaging] update variants
* Pull request: Add quirk and disable SBR on Gen5 ports (LP: #2107509)
- NVIDIA: SAUCE: MEDIATEK: usb: host: xhci-plat: support usb3 bulks stream
low
power
- NVIDIA: SAUCE: MEDIATEK: PCI: disable SBR on Gen5 ports
* Pull-request: Patches to support CPPCv3 Autonomous Frequency Control
(LP: #2102685)
- NVIDIA: SAUCE: ACPI: CPPC: rename cppc_perf apis
- NVIDIA: SAUCE: ACPI: CPPC: add read perf ctrls api
- NVIDIA: SAUCE: ACPI: CPPC: expand macro to create store acpi_cppc sysfs
node
- NVIDIA: SAUCE: ACPI: CPPC: support updating epp, auto_sel and
{min|max_perf}
from sysfs
- NVIDIA: SAUCE: Documentation: ACPI: add autonomous mode ctrls info in
cppc_sysfs.rst
- NVIDIA: SAUCE: cpufreq: CPPC: Add cppc_cpufreq_epp instance for Autonomous
mode
* Apply cherrypick of upstream patch to recognize PNP0D15 usb3 device
(LP: #2097644)
- NVIDIA: SAUCE: Fixes the kernel boot issues due to xhci mem errors
* Apply backport of upstream commit to enable Realtek Bluetooth module
(LP: #2096882)
- NVIDIA: SAUCE: Adds MT7925 BT devices
* Apply SAUCE patch to enable 8250 serial device (LP: #2096888)
- NVIDIA: SAUCE: serial: 8250_mtk: Add ACPI support
* Backport: "Add support for AArch64 AMUv1-based average freq" Series
(LP: #2100032)
- arch_topology: init capacity_freq_ref to 0
- cpufreq: Allow arch_freq_get_on_cpu to return an error
- cpufreq: Introduce an optional cpuinfo_avg_freq sysfs entry
- arm64: Provide an AMU-based version of arch_freq_get_on_cpu
- arm64: Update AMU-based freq scale factor on entering idle
- arm64: Utilize for_each_cpu_wrap for reference lookup
- NVIDIA: [Config] set CONFIG_CPUFREQ_ARCH_CUR_FREQ=y for x86
* Pull request to enable GPU passthrough for CUDA (LP: #2095028)
- NVIDIA: SAUCE: iommu/arm-smmu-v3: Implement
arm_smmu_get_msi_mapping_domain
- NVIDIA: SAUCE: iommu/dma: Support MSIs through nested domains
- NVIDIA: SAUCE: KVM: arm64: determine memory type from VMA
- NVIDIA: SAUCE: WAR: Expose PCI PASID capability to userspace
- NVIDIA: SAUCE: WAR: iommufd/pages: Bypass PFNMAP
- NVIDIA: SAUCE: [Config] nvidia-6.14: Update annotations for Grace I/O
virtualization
- NVIDIA: SAUCE: Revert "iommu/tegra241-cmdqv: Do not allocate vcmdq until
dma_set_mask_and_coherent"
- iommu/tegra241-cmdqv: Do not allocate vcmdq until
dma_set_mask_and_coherent
* PCI resources unassigned with Bluefield3 switch (LP: #2094821)
- NVIDIA: SAUCE: PCI: Use downstream bridges for distributing resources
* MANA: include driver fixes and enable module on ARM64 (LP: #2084598)
- [Config] nvidia-6.11: Enable MANA configs on x86 and arm64
- [Packaging] nvidia-6.11: Include mana.ko in linux-modules-ABIVER package
* Apply patch to set CONFIG_EFI_CAPSULE_LOADER=y for arm64 (LP: #2067111)
- NVIDIA: [Config] EFI: set CAPSULE_LOADER=y for arm64
* Export kernel symbols required for NVIDIA GDS (LP: #2068544)
- NVIDIA: SAUCE: NFS: Export nvfs register and unregister functions as GPL
- NVIDIA: SAUCE: NVMe/NVMeoF: Export nvfs register and unregister functions
as
GPL
* linux-nvidia-6.5_6.5.0-1014.14 breaks with earlier BIOS release, and
modeset/resolutions are wrong (LP: #2061930) // Blacklist coresight_etm4x
(LP: #2067106)
- [Packaging] blacklist coresight_etm4x
* backport arm64 THP improvements from 6.9 (LP: #2059316)
- NVIDIA: [Config] arm64: ARM64_CONTPTE=y
* Enable GDS in the 6.8 based linux-nvidia kernel (LP: #2059814)
- NVIDIA: SAUCE: Patch NFS driver to support GDS with 6.8 Kernel
- NVIDIA: SAUCE: NVMe/MVMEeOF: Patch NVMe/NVMeOF driver to support GDS on
Linux 6.8 Kernel
- NVIDIA: [Config] Add nvidia-fs build dependencies
* Reapply the linux-nvidia kernel config options from the 5.15 and 6.5 kernels
(LP: #2060327)
- NVIDIA: [Config]: Disable the NOUVEAU driver which is not used with
-nvidia
kernels
- NVIDIA: [Config]: Adding CORESIGHT and ARM64_ERRATUM configs to
annotations
-- Jacob Martin <jacob.mar...@canonical.com> Fri, 30 May 2025 19:09:52
-0500
--
You received this bug notification because you are a member of Kernel
Packages, which is subscribed to linux-nvidia-6.11 in Ubuntu.
https://bugs.launchpad.net/bugs/2110750
Title:
UBSAN: shift-out-of-bounds arm-smmu-v3.c
Status in linux-nvidia-6.11 package in Ubuntu:
Invalid
Status in linux-nvidia-6.14 package in Ubuntu:
Invalid
Status in linux-nvidia-6.11 source package in Noble:
Fix Released
Status in linux-nvidia-6.14 source package in Noble:
Fix Released
Bug description:
UBSAN: shift-out-bounds in drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
When running cuda test that exercise SVA it is possible to see this
stack trace:
[ 7825.373952] ------------[ cut here ]------------
[ 7825.373960] UBSAN: shift-out-of-bounds in
drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c:2213:20
[ 7825.382443] shift exponent 76 is too large for 64-bit type 'long unsigned
int'
[ 7825.389619] CPU: 5 UID: 0 PID: 96250 Comm: ats_memory_acce Tainted: P
[ 7825.389624] Tainted: [P]=PROPRIETARY_MODULE, [O]=OOT_MODULE,
[E]=UNSIGNED_MODULE
[ 7825.389626] Hardware name: Nvidia
[ 7825.389628] Call trace:
[ 7825.389630] show_stack+0x20/0x48 (C)
[ 7825.389639] dump_stack_lvl+0x80/0x130
[ 7825.389647] dump_stack+0x18/0x30
[ 7825.389650] __ubsan_handle_shift_out_of_bounds+0x144/0x250
[ 7825.389658] __arm_smmu_tlb_inv_range+0x240/0x268
[ 7825.389665] arm_smmu_tlb_inv_range_asid+0x6c/0xd0
[ 7825.389669] arm_smmu_mm_arch_invalidate_secondary_tlbs+0xb8/0xc8
[ 7825.389674] __mmu_notifier_arch_invalidate_secondary_tlbs+0x70/0xd0
[ 7825.389683] ptep_clear_flush+0x16c/0x180
[ 7825.389689] do_wp_page+0x4f0/0xfe8
[ 7825.389692] __handle_mm_fault+0x580/0xa60
[ 7825.389694] handle_mm_fault+0xf4/0x340
[ 7825.389697] do_page_fault+0x11c/0x678
[ 7825.389701] do_mem_abort+0x4c/0xc8
[ 7825.389704] el0_da+0x48/0x180
[ 7825.389707] el0t_64_sync_handler+0x154/0x188
[ 7825.389709] el0t_64_sync+0x1b0/0x1b8
[ 7825.389712] ---[ end trace ]---
The patch o fix this issue:
https://web.git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/patch/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3-sva.c?id=12f78021973ae422564b234136c702a305932d73
We will want this patch for 6.11 and 6.14 kernel.
To manage notifications about this bug go to:
https://bugs.launchpad.net/ubuntu/+source/linux-nvidia-6.11/+bug/2110750/+subscriptions
--
Mailing list: https://launchpad.net/~kernel-packages
Post to : kernel-packages@lists.launchpad.net
Unsubscribe : https://launchpad.net/~kernel-packages
More help : https://help.launchpad.net/ListHelp
