Skip to content

[NV TRT RTX EP] Reconfigure memory arena to grow with power of 2 #25800

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
jywu-msft merged 1 commit into from
Aug 22, 2025
Merged

[NV TRT RTX EP] Reconfigure memory arena to grow with power of 2 #25800

Changes from all commits
Commits
Show all changes
1 commit
Select commit
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
6 changes: 3 additions & 3 deletions onnxruntime/core/providers/nv_tensorrt_rtx/nv_execution_provider.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1093,7 +1093,7 @@ void NvExecutionProvider::IncrementRegularRunCountBeforeGraphCapture() {
}

std::vector<AllocatorPtr> NvExecutionProvider::CreatePreferredAllocators() {
OrtArenaCfg arena_cfg(0, static_cast<int>(ArenaExtendStrategy::kSameAsRequested),
OrtArenaCfg arena_cfg(0, static_cast<int>(ArenaExtendStrategy::kNextPowerOfTwo),
-1, -1, -1, -1);
AllocatorCreationInfo default_memory_info(
[](OrtDevice::DeviceId device_id) { return std::make_unique<CUDAAllocator>(device_id, CUDA); },
Expand Down Expand Up @@ -2650,7 +2650,7 @@ Status NvExecutionProvider::CreateNodeComputeInfoFromGraph(const GraphViewer& gr
}
if (trt_state->context_memory_size != mem_size) {
LOGS_DEFAULT(INFO) << "[NvTensorRTRTX EP] A new context memory was allocated with size " << mem_size;
trt_state->context_memory = IAllocator::MakeUniquePtrFromOrtAllocator<void>(alloc, mem_size, false /*use_reserve*/);
trt_state->context_memory = IAllocator::MakeUniquePtrFromOrtAllocator<void>(alloc, mem_size, true /*use_reserve*/);
trt_state->context_memory_size = mem_size;
trt_context->setDeviceMemoryV2(trt_state->context_memory.get(), mem_size);
}
Expand Down Expand Up @@ -2963,7 +2963,7 @@ Status NvExecutionProvider::CreateNodeComputeInfoFromPrecompiledEngine(const Gra
}
if (trt_state->context_memory_size != mem_size) {
LOGS_DEFAULT(INFO) << "[NvTensorRTRTX EP] A new context memory was allocated with size " << mem_size;
trt_state->context_memory = IAllocator::MakeUniquePtrFromOrtAllocator<void>(alloc, mem_size, false /*use_reserve*/);
trt_state->context_memory = IAllocator::MakeUniquePtrFromOrtAllocator<void>(alloc, mem_size, true /*use_reserve*/);
// trt_state->context_memory = IAllocator::MakeUniquePtr<void>(alloc, mem_size, false /*use_reserve*/, stream);
trt_state->context_memory_size = mem_size;
trt_context->setDeviceMemoryV2(trt_state->context_memory.get(), mem_size);
Expand Down
Loading