APPLIED: [SRU][O:linux-aws][PULL] Backport patches to support NVIDIA GB200

Magali Lemes magali.lemes at canonical.com
Thu Mar 27 13:28:02 UTC 2025


On 17/03/2025 15:42, Magali Lemes wrote:
> BugLink: https://bugs.launchpad.net/bugs/2101185
> 
> [Impact]
> AWS requested the following patchsets to support GB200 to be backported
> to 6.8 kernels and newer:
> #1: Update SMMUv3 to the modern iommu API (part 2b/3) (https:// 
> lore.kernel.org/linux-iommu/0-v9-5cd718286059+79186- 
> smmuv3_newapi_p2b_jgg at nvidia.com/)
> #2: Add Tegra241 (Grace) CMDQV Support (part 1/2) (https:// 
> lore.kernel.org/linux-iommu/cover.1724970714.git.nicolinc at nvidia.com/)
> #3: iommu/tegra241-cmdqv: Fix alignment failure at max_n_shift (https:// 
> lore.kernel.org/all/20241111030226.1940737-1-nicolinc at nvidia.com/)
> 
> [Fix]
> - Patchset #1 is already included in the o:aws tree.
> - All patches from patchset #2 are clean cherry-picks. Fixes to commit
> 918eb5c856f6 ("iommu/arm-smmu-v3: Add in-kernel support for NVIDIA 
> Tegra241 (Grace) CMDQV"), from patchset 2, which were also clean cherry- 
> picks, were added.
> - Patchset #3's only patch and one fix to it were clean cherry-picks.
> 
> [Test Case]
> Compile and boot tested.
> Tested by AWS.
> 
> [Where problems could occur]
> These patches introduce significant refactoring and features to the
> SMMUv3 driver. Additionally, support to Tegra241's CMDQ-Virtualization
> is being added.
> 
> [Other info]
> SF #00404773
> 
> ------------------------------------------------------------------------
> 
> The following changes since commit 
> f7dffa22504c9b6506d9fbe3368fffaad1469562:
> 
>    UBUNTU: Ubuntu-aws-6.11.0-1011.12 (2025-02-25 12:47:55 -0500)
> 
> are available in the Git repository at:
> 
> 
> git://git.launchpad.net/~magalilemes/ubuntu/+source/linux-aws/+git/ 
> oracular gb200
> 
> for you to fetch changes up to 225b718bf5084356529e51b91828eae0f93e60cf:
> 
>    iommu/tegra241-cmdqv: Read SMMU IDR1.CMDQS instead of hardcoding 
> (2025-03-17 13:32:30 -0300)
> 
> ----------------------------------------------------------------
> Dan Carpenter (1):
>        iommu/tegra241-cmdqv: Fix ioremap() error handling in probe()
> 
> Jason Gunthorpe (1):
>        iommu/arm-smmu-v3: Add struct arm_smmu_impl_ops
> 
> Luis Claudio R. Goncalves (1):
>        iommu/tegra241-cmdqv: do not use smp_processor_id in preemptible 
> context
> 
> Magali Lemes (1):
>        UBUNTU: [Config] updateconfigs to enable CONFIG_TEGRA241_CMDQV
> 
> Nate Watterson (1):
>        iommu/arm-smmu-v3: Add in-kernel support for NVIDIA Tegra241 
> (Grace) CMDQV
> 
> Nicolin Chen (14):
>        iommu/arm-smmu-v3: Issue a batch of commands to the same cmdq
>        iommu/arm-smmu-v3: Pass in cmdq pointer to 
> arm_smmu_cmdq_build_sync_cmd
>        iommu/arm-smmu-v3: Pass in cmdq pointer to arm_smmu_cmdq_init
>        iommu/arm-smmu-v3: Make symbols public for CONFIG_TEGRA241_CMDQV
>        iommu/arm-smmu-v3: Add ARM_SMMU_OPT_TEGRA241_CMDQV
>        iommu/arm-smmu-v3: Add acpi_smmu_iort_probe_model for impl
>        iommu/arm-smmu-v3: Start a new batch if new command is not supported
>        iommu/tegra241-cmdqv: Limit CMDs for VCMDQs of a guest owned VINTF
>        iommu/tegra241-cmdqv: Fix -Wformat-truncation warnings in 
> lvcmdq_error_header
>        iommu/tegra241-cmdqv: Drop static at local variable
>        iommu/tegra241-cmdqv: Do not allocate vcmdq until 
> dma_set_mask_and_coherent
>        iommu/tegra241-cmdqv: Staticize cmdqv_debugfs_dir
>        iommu/tegra241-cmdqv: Fix alignment failure at max_n_shift
>        iommu/tegra241-cmdqv: Read SMMU IDR1.CMDQS instead of hardcoding
> 
>   MAINTAINERS                                    |   1 +
>   debian.aws/config/annotations                  |   1 +
>   drivers/iommu/Kconfig                          |  11 +++
>   drivers/iommu/arm/arm-smmu-v3/Makefile         |   1 +
>   drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c    | 212 +++++++++++++++++ 
> +++++++++++++++++---------
>   drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.h    |  41 +++++++++
>   drivers/iommu/arm/arm-smmu-v3/tegra241-cmdqv.c | 912 +++++++++++++++++ 
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 
> ++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++ 
> +++++++++++++++++++++
>   7 files changed, 1135 insertions(+), 44 deletions(-)
>   create mode 100644 drivers/iommu/arm/arm-smmu-v3/tegra241-cmdqv.c
> 

Applied to oracular:linux-aws/master-next. Thanks!


Magali



More information about the kernel-team mailing list