Cherry pick jetson enablement from 2.8 release branch to main

[lanluo-nvidia]

Description

Please include a summary of the change and which issue is fixed. Please also include relevant motivation and context. List any dependencies that are required for this change.

Fixes # (issue)

Type of change

Please delete options that are not relevant and/or add your own.

• Bug fix (non-breaking change which fixes an issue)
• New feature (non-breaking change which adds functionality)
• Breaking change (fix or feature that would cause existing functionality to not work as expected)
• This change requires a documentation update

Checklist:

• My code follows the style guidelines of this project (You can use the linters)
• I have performed a self-review of my own code
• I have commented my code, particularly in hard-to-understand areas and hacks
• I have made corresponding changes to the documentation
• I have added tests to verify my fix or my feature
• New and existing unit tests pass locally with my changes
• I have added the relevant labels to my PR in so that relevant reviewers are notified

[github-actions]

There are some changes that do not conform to Python style guidelines:

--- /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/conversion/custom_ops_converters.py	2025-08-08 06:50:04.271977+00:00
+++ /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/conversion/custom_ops_converters.py	2025-08-08 06:50:29.351625+00:00
@@ -19,10 +19,11 @@
if load_tensorrt_llm():
    from torch_tensorrt.dynamo.lowering.passes.fuse_distributed_ops import (
        tensorrt_fused_nccl_all_gather_op,
        tensorrt_fused_nccl_reduce_scatter_op,
    )
+
    @dynamo_tensorrt_converter(tensorrt_fused_nccl_all_gather_op)
    def fused_nccl_gather(
        ctx: ConversionContext,
        target: Target,
        args: Tuple[Argument, ...],
--- /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/lowering/passes/_aten_lowering_pass.py	2025-08-08 06:50:04.274977+00:00
+++ /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/lowering/passes/_aten_lowering_pass.py	2025-08-08 06:50:30.337911+00:00
@@ -33,10 +33,11 @@
    remove_detach,
]

if not is_tegra_platform():
    from .fuse_distributed_ops import fuse_distributed_ops
+
    post_lowering_pass_list.append(fuse_distributed_ops)

ATEN_POST_LOWERING_PASSES = DynamoPassManager.build_from_passlist(
    post_lowering_pass_list
)
--- /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/conversion/aten_ops_converters.py	2025-08-08 06:50:04.271977+00:00
+++ /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/conversion/aten_ops_converters.py	2025-08-08 06:50:32.115974+00:00
@@ -617,10 +617,11 @@
            args[0],
            args[1],
            args[2],
            args[3],
        )
+

if is_tensorrt_version_supported("10.8.0"):
    try:
        import modelopt.torch.quantization as mtq  # noqa: F401

[github-actions]

There are some changes that do not conform to Python style guidelines:

--- /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/conversion/custom_ops_converters.py	2025-08-08 06:55:35.012666+00:00
+++ /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/conversion/custom_ops_converters.py	2025-08-08 06:56:00.306350+00:00
@@ -19,10 +19,11 @@
if load_tensorrt_llm():
    from torch_tensorrt.dynamo.lowering.passes.fuse_distributed_ops import (
        tensorrt_fused_nccl_all_gather_op,
        tensorrt_fused_nccl_reduce_scatter_op,
    )
+
    @dynamo_tensorrt_converter(tensorrt_fused_nccl_all_gather_op)
    def fused_nccl_gather(
        ctx: ConversionContext,
        target: Target,
        args: Tuple[Argument, ...],
--- /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/lowering/passes/_aten_lowering_pass.py	2025-08-08 06:55:35.015667+00:00
+++ /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/lowering/passes/_aten_lowering_pass.py	2025-08-08 06:56:01.320835+00:00
@@ -33,10 +33,11 @@
    remove_detach,
]

if not is_tegra_platform():
    from .fuse_distributed_ops import fuse_distributed_ops
+
    post_lowering_pass_list.append(fuse_distributed_ops)

ATEN_POST_LOWERING_PASSES = DynamoPassManager.build_from_passlist(
    post_lowering_pass_list
)
--- /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/conversion/aten_ops_converters.py	2025-08-08 06:55:35.012666+00:00
+++ /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/conversion/aten_ops_converters.py	2025-08-08 06:56:03.108230+00:00
@@ -617,10 +617,11 @@
            args[0],
            args[1],
            args[2],
            args[3],
        )
+

if is_tensorrt_version_supported("10.8.0"):
    try:
        import modelopt.torch.quantization as mtq  # noqa: F401

[github-actions]

There are some changes that do not conform to Python style guidelines:

--- /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/conversion/custom_ops_converters.py	2025-08-08 07:19:01.667895+00:00
+++ /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/conversion/custom_ops_converters.py	2025-08-08 07:19:27.514217+00:00
@@ -19,10 +19,11 @@
if load_tensorrt_llm():
    from torch_tensorrt.dynamo.lowering.passes.fuse_distributed_ops import (
        tensorrt_fused_nccl_all_gather_op,
        tensorrt_fused_nccl_reduce_scatter_op,
    )
+
    @dynamo_tensorrt_converter(tensorrt_fused_nccl_all_gather_op)
    def fused_nccl_gather(
        ctx: ConversionContext,
        target: Target,
        args: Tuple[Argument, ...],
--- /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/lowering/passes/_aten_lowering_pass.py	2025-08-08 07:19:01.670895+00:00
+++ /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/lowering/passes/_aten_lowering_pass.py	2025-08-08 07:19:28.513808+00:00
@@ -33,10 +33,11 @@
    remove_detach,
]

if not is_tegra_platform():
    from .fuse_distributed_ops import fuse_distributed_ops
+
    post_lowering_pass_list.append(fuse_distributed_ops)

ATEN_POST_LOWERING_PASSES = DynamoPassManager.build_from_passlist(
    post_lowering_pass_list
)
--- /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/conversion/aten_ops_converters.py	2025-08-08 07:19:01.667895+00:00
+++ /home/runner/work/TensorRT/TensorRT/py/torch_tensorrt/dynamo/conversion/aten_ops_converters.py	2025-08-08 07:19:30.393872+00:00
@@ -617,10 +617,11 @@
            args[0],
            args[1],
            args[2],
            args[3],
        )
+

if is_tensorrt_version_supported("10.8.0"):
    try:
        import modelopt.torch.quantization as mtq  # noqa: F401

[pytorch-bot]

No ciflow labels are configured for this repo.
For information on how to enable CIFlow bot see this wiki