Struct EpistemicGpuFinalTupleMaterializationTrace 

pub struct EpistemicGpuFinalTupleMaterializationTrace {

    pub output_column_count: usize,
    pub output_row_capacity: usize,
    pub tuple_bytes_capacity: usize,
    pub output_row_count_device_reads: u32,
    pub model_membership_bytes_checked: usize,
    pub bounded_model_slots_per_reduction: usize,
    pub row_specific_membership_row_capacity: usize,
    pub row_filter_row_capacity_outside_model_slot_window: usize,
    pub world_view_slots_checked: usize,
    pub row_filter_count: usize,
    pub negated_row_filter_count: usize,
    pub final_row_count_device_writes: u32,
    pub kernel_launches: u32,
    pub host_write_ops: u32,
    pub kernel_timing: EpistemicGpuKernelTimingTrace,
}

Trace proving final query tuples were materialized into a device-resident buffer.

Fields

output_column_count: usize

Number of output columns copied into the final device buffer.

output_row_capacity: usize

Row capacity of the final output buffer.

tuple_bytes_capacity: usize

Device tuple bytes covered by the materialization kernels.

output_row_count_device_reads: u32

Device output row-count scalars read by the kernels.

model_membership_bytes_checked: usize

Model-membership bytes checked by the kernels before tuple materialization.

bounded_model_slots_per_reduction: usize

Bounded model slots available per reduction during final tuple materialization.

row_specific_membership_row_capacity: usize

Output row capacity that can be checked against row-specific model slots.

row_filter_row_capacity_outside_model_slot_window: usize

Output row capacity beyond the bounded model-slot window.

world_view_slots_checked: usize

World-view slots checked by the kernels before tuple materialization.

row_filter_count: usize

Variable-bound tuple row filters applied by the final-row map kernel.

negated_row_filter_count: usize

Negated variable-bound tuple row filters applied by the final-row map kernel.

final_row_count_device_writes: u32

Device final row-count scalars written by the kernels.

kernel_launches: u32

Final tuple materialization kernel launches.

host_write_ops: u32

Host writes used by final tuple materialization. Accepted execution requires zero.

kernel_timing: EpistemicGpuKernelTimingTrace

CUDA-event timing for the launched kernel batch.

Implementations

impl EpistemicGpuFinalTupleMaterializationTrace

pub fn for_counts( output_column_count: usize, output_row_capacity: usize, tuple_bytes_capacity: usize, literal_count: usize, candidate_count: usize, reduction_count: usize, models_per_reduction: usize, ) -> Result<Self>

Build a final tuple materialization trace for a device-side output buffer.

pub const fn with_kernel_timing( self, kernel_timing: EpistemicGpuKernelTimingTrace, ) -> Self

Attach CUDA-event timing captured by the runtime launch path.

pub fn with_row_filter_counts( self, row_filter_count: usize, negated_row_filter_count: usize, ) -> Result<Self>

Attach final-row filter metadata captured before launching the row-map kernel.

pub fn require_row_filter_materialization_evidence( &self, construct: &str, final_output_rows: usize, ) -> Result<()>

Require GPU evidence that row-filtered tuple output fits the validated coverage window.

Trait Implementations

impl Clone for EpistemicGpuFinalTupleMaterializationTrace

fn clone(&self) -> EpistemicGpuFinalTupleMaterializationTrace

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Copy for EpistemicGpuFinalTupleMaterializationTrace

impl Debug for EpistemicGpuFinalTupleMaterializationTrace

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Eq for EpistemicGpuFinalTupleMaterializationTrace

impl PartialEq for EpistemicGpuFinalTupleMaterializationTrace

fn eq(&self, other: &EpistemicGpuFinalTupleMaterializationTrace) -> bool

Tests for self and other values to be equal, and is used by ==.

fn ne(&self, other: &Rhs) -> bool

Tests for !=. The default implementation is almost always sufficient, and should not be overridden without very good reason.

impl StructuralPartialEq for EpistemicGpuFinalTupleMaterializationTrace