ProcessMlaOptions Struct

process-MLA execution options. More...

Declaration

struct simaai::neat::ProcessMlaOptions { ... }

Included Headers

#include <GraphOptions.h>

Public Member Attributes Index

bool	async = true
	Enable the safe async processmla submit/emit path. More...
int	output_pool_buffers = 0
	Optional processmla output pool size override. More...
bool	defer_output_invalidate = true
	Compatibility field: must remain true. More...

Description

process-MLA execution options.

Definition at line 267 of file GraphOptions.h.

Public Member Attributes

async

bool simaai::neat::ProcessMlaOptions::async = true

Enable the safe async processmla submit/emit path.

Default true uses the optimized prepared async lane when the plugin/stage is eligible; set false, or use the plugin/env kill switches, to force the synchronous fallback.

Definition at line 271 of file GraphOptions.h.

defer_output_invalidate

bool simaai::neat::ProcessMlaOptions::defer_output_invalidate = true

Compatibility field: must remain true.

Scheduler always defers CPU visibility to the actual consumer through Runner prepared access. False is rejected; graph topology no longer selects producer-side READ epochs.

Definition at line 282 of file GraphOptions.h.

output_pool_buffers

int simaai::neat::ProcessMlaOptions::output_pool_buffers = 0

Optional processmla output pool size override.

A value <= 0 leaves the runtime default in place. The framework runtime default is 4, matching model-managed CVU/MLA buffering and avoiding artificial backpressure while downstream stages still hold previous tensor-set outputs.

Definition at line 277 of file GraphOptions.h.

---

The documentation for this struct was generated from the following file:

• GraphOptions.h

---

Generated via doxygen2docusaurus 2.0.0 by Doxygen 1.15.0.