Skip to content

[https://nvbugs/5702793][fix] Fix view operation on uncontiguous tensor #10147

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Merged
shuyixiong merged 2 commits into from
Dec 21, 2025
+6 −6
Merged

[https://nvbugs/5702793][fix] Fix view operation on uncontiguous tensor #10147

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
ab87c93
Fix view operation on uncontiguous tensor
shuyixiong Dec 19, 2025
674826d
Unwaive test
shuyixiong Dec 20, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 6 additions & 4 deletions tensorrt_llm/_torch/modules/fused_moe/quantization.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -2101,9 +2101,9 @@ def load_expert_w3_w1_weight_scale_nvfp4(
TensorParallelMode.COLUMN,
device=device)

cast_w3_weight_scale = w3_weight_scale.view(
cast_w3_weight_scale = w3_weight_scale.contiguous().view(
dst_w3_w1_weight_scale.dtype)
cast_w1_weight_scale = w1_weight_scale.view(
cast_w1_weight_scale = w1_weight_scale.contiguous().view(
dst_w3_w1_weight_scale.dtype)
cast_w31_weight_scale = torch.cat(
[cast_w3_weight_scale, cast_w1_weight_scale], dim=0)
Expand Down Expand Up @@ -2165,8 +2165,10 @@ def load_expert_w3_w1_weight(self, module: torch.nn.Module,
TensorParallelMode.COLUMN,
device=device)

cast_w1_weight_shard = w1_weight_shard.view(dst_w3_w1_weight.dtype)
cast_w3_weight_shard = w3_weight_shard.view(dst_w3_w1_weight.dtype)
cast_w1_weight_shard = w1_weight_shard.contiguous().view(
dst_w3_w1_weight.dtype)
cast_w3_weight_shard = w3_weight_shard.contiguous().view(
dst_w3_w1_weight.dtype)
cast_w31_weight_shard = torch.cat(
[cast_w3_weight_shard, cast_w1_weight_shard], dim=0)
cast_w31_weight_shard = self._maybe_padding_shape(
Expand Down
2 changes: 0 additions & 2 deletions tests/integration/test_lists/waives.txt
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -476,8 +476,6 @@ accuracy/test_cli_flow.py::TestPhi3Small128kInstruct::test_auto_dtype SKIP (http
accuracy/test_cli_flow.py::TestPhi3_5MiniInstruct::test_auto_dtype SKIP (https://nvbugs/5744293)
unittest/_torch/auto_deploy/unit/singlegpu/models/test_llama4_vlm_patch.py::test_build_run_llama4_vlm SKIP (https://nvbugs/5747878)
accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_nvfp4_4gpus_online_eplb[fp8kv=True-moe_backend=TRTLLM] SKIP (https://nvbugs/5740377)
accuracy/test_llm_api_pytorch.py::TestLlama4ScoutInstruct::test_fp4_chunked_prefill[tp4ep4-cuda_graph=True] SKIP (https://nvbugs/5702793)
accuracy/test_llm_api_pytorch.py::TestLlama4ScoutInstruct::test_fp4[tp4-cuda_graph=True] SKIP (https://nvbugs/5702793)
disaggregated/test_auto_scaling.py::test_minimal_instances[etcd-round_robin] SKIP (https://nvbugs/5748564)
unittest/llmapi/apps/test_disagg_serving_perf_metrics.py SKIP (https://nvbugs/5752516)
unittest/_torch/attention/test_trtllm_flashinfer_symbol_collision.py::test_flashinfer_fused_moe_matches_torch_moe SKIP (https://nvbugs/5752521)
Expand Down