Add cuda graph implementation for NV TRT RTX EP #25787
TRT RTX workspace allocation using ORT Arena allocator
8f4dae0f
Reduce CPU overhead of TRT-RTX EP's compute function
37757a39
Adding ORT framework side for CUDA graph in Nv TensorRT RTX
1efc1b5f
Enabled CUDA Graph for NV TRT RTX EP
d922edcb
Comments addressed
ce054e49
Comments addressed
d4e737a7
Added ORT framework support functions and removed cuda_graph file cha…
fd2f0a67
fixed comments
d6895747
removing this chnage as it is breaking the functional validation and …
c141b5a9
Fixed gauravs comments
9a4f3932
Fixed flag and context creation
f8e7ca3d
Fixed flag name
50f905f2
First commit Fix
2ce8f1ce
First Initial fix commit
49179b4a
Fix for CG with rebase
9a426837
Fixed Max and Ishwars comments
68691511
Post-rebase change
32e6d42d
Fixed comments and rebase error
d293b0a8
Fixed comment and reverted cuda_graph_enable flag
9e477bc9
Fixed comments and recapture break bug
45551984
Fixed comments for false addition and new comments
c27ffdac
Fixed sync and cuda_graph flag for ORTGenAI
b96d7a65
Fixed copilot comments
a75ac43b
Fixed clang-format errors
49a8b7f8
The Replay function CUDA EP Commnet
a376b3c5
Fixed the function definition style
127e1add
Fixed the rebase error
c4fd3932
Merge branch 'main' into umang/cuda_graph_msr_main
a91ea79d
chilo-ms
approved these changes
on 2025-08-26
chilo-ms
merged
16ae99ed
into main 196 days ago
snnn
removed release:1.23.0
Assignees
No one assigned
Login to write a write a comment.
Login via GitHub