Struct EpistemicGpuModelMembershipTrace 

pub struct EpistemicGpuModelMembershipTrace {

    pub literal_count: usize,
    pub candidates_checked: usize,
    pub reduction_count: usize,
    pub models_per_reduction: usize,
    pub model_membership_bytes_written: usize,
    pub output_row_count_device_reads: u32,
    pub tuple_source_row_count_device_reads: u32,
    pub tuple_source_key_column_device_reads: u32,
    pub rejection_reason_slots_checked: usize,
    pub membership_source: EpistemicGpuModelMembershipSource,
    pub kernel_launches: u32,
    pub host_write_ops: u32,
    pub kernel_timing: EpistemicGpuKernelTimingTrace,
}

Trace proving model-membership staging was performed by a GPU kernel.

Fields

literal_count: usize

Number of epistemic literals represented per candidate/model.

candidates_checked: usize

Number of candidate rows checked on device.

reduction_count: usize

Number of reduced-program summaries represented in the membership layout.

models_per_reduction: usize

Maximum models represented per reduction.

model_membership_bytes_written: usize

Model-membership bytes written by the kernel.

output_row_count_device_reads: u32

Device output row-count scalars read by the kernel.

tuple_source_row_count_device_reads: u32

Device tuple-source row-count scalars read by the kernel.

tuple_source_key_column_device_reads: u32

Device tuple-key columns read by tuple-source membership kernels.

rejection_reason_slots_checked: usize

Rejection-reason slots checked by the kernel.

membership_source: EpistemicGpuModelMembershipSource

Source used to populate model-membership bytes.

kernel_launches: u32

Model-membership staging kernel launches.

host_write_ops: u32

Host writes used by model-membership staging. Accepted execution requires zero.

kernel_timing: EpistemicGpuKernelTimingTrace

CUDA-event timing for the launched kernel.

Implementations

impl EpistemicGpuModelMembershipTrace

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

Build a model-membership trace for a bounded device launch.

pub fn for_stable_model_tuple_sources( literal_count: usize, candidate_count: usize, reduction_count: usize, models_per_reduction: usize, tuple_source_count: usize, ) -> Result<Self>

Build a model-membership trace backed by reduced stable-model tuple sources.

pub fn for_stable_model_tuple_sources_with_key_columns( literal_count: usize, candidate_count: usize, reduction_count: usize, models_per_reduction: usize, tuple_source_count: usize, tuple_source_key_column_count: usize, ) -> Result<Self>

Build a model-membership trace backed by tuple sources and key columns.

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

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

pub fn require_stable_model_tuple_source(&self) -> Result<()>

Require semantic stable-model tuple membership before accepting execution.

pub fn require_planned_tuple_key_column_reads( &self, expected_key_column_reads: usize, ) -> Result<()>

Require the tuple-key device reads planned for this model-membership trace.

Trait Implementations

impl Clone for EpistemicGpuModelMembershipTrace

fn clone(&self) -> EpistemicGpuModelMembershipTrace

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Copy for EpistemicGpuModelMembershipTrace

impl Debug for EpistemicGpuModelMembershipTrace

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

Formats the value using the given formatter.

impl Eq for EpistemicGpuModelMembershipTrace

impl PartialEq for EpistemicGpuModelMembershipTrace

fn eq(&self, other: &EpistemicGpuModelMembershipTrace) -> 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 EpistemicGpuModelMembershipTrace