//! Comprehensive performance validation for Foxhunt HFT system
//!
//! This benchmark validates the claimed performance metrics:
//! - 14ns RDTSC timing precision
//! - Sub-50μs latency claims
//! - SIMD/AVX2 optimizations
//! - Lock-free data structure performance

use criterion::{black_box, criterion_group, criterion_main, BenchmarkId, Criterion};
use std::time::{Duration, Instant};

// Core performance modules
use foxhunt_core::lockfree::{message_types, HftMessage, SharedMemoryChannel};
use foxhunt_core::simd::{AlignedPrices, AlignedVolumes, SafeSimdDispatcher, SimdLevel};
use foxhunt_core::timing::{calibrate_tsc, HardwareTimestamp, LatencyMeasurement};

fn benchmark_rdtsc_timing(c: &mut Criterion) {
    let mut group = c.benchmark_group("RDTSC Timing");

    // Try to calibrate TSC
    let _tsc_freq = calibrate_tsc();

    group.bench_function("timestamp_capture", |b| {
        b.iter(|| {
            let ts = HardwareTimestamp::now();
            black_box(ts);
        });
    });

    group.bench_function("latency_calculation", |b| {
        let ts1 = HardwareTimestamp::now();
        std::thread::sleep(Duration::from_nanos(100)); // Small delay
        let ts2 = HardwareTimestamp::now();

        b.iter(|| {
            let latency = ts2.latency_ns(&ts1);
            black_box(latency);
        });
    });

    group.bench_function("measurement_overhead", |b| {
        b.iter(|| {
            let mut measurement = LatencyMeasurement::start();
            let _latency = measurement.finish();
            black_box(_latency);
        });
    });

    group.finish();
}

fn benchmark_simd_operations(c: &mut Criterion) {
    let mut group = c.benchmark_group("SIMD Operations");

    let dispatcher = SafeSimdDispatcher::new();
    println!("SIMD Level: {}", dispatcher.simd_level());

    // Test data
    let prices: Vec<f64> = (0..1000).map(|i| 100.0 + i as f64 * 0.01).collect();
    let volumes: Vec<f64> = (0..1000).map(|i| 1000.0 + i as f64).collect();

    // Test different data sizes
    for size in [100, 500, 1000].iter() {
        let test_prices = &prices[..*size];
        let test_volumes = &volumes[..*size];

        group.bench_with_input(BenchmarkId::new("vwap_calculation", size), size, |b, _| {
            let adaptive_ops = dispatcher.create_adaptive_price_ops();
            b.iter(|| {
                let vwap = adaptive_ops.calculate_vwap(test_prices, test_volumes);
                black_box(vwap);
            });
        });

        // Compare SIMD vs scalar performance
        if dispatcher.simd_level() >= SimdLevel::AVX2 {
            group.bench_with_input(
                BenchmarkId::new("vwap_simd_vs_scalar", size),
                size,
                |b, _| {
                    b.iter(|| {
                        // SIMD calculation
                        let adaptive_ops = dispatcher.create_adaptive_price_ops();
                        let simd_vwap = adaptive_ops.calculate_vwap(test_prices, test_volumes);

                        // Scalar calculation for comparison
                        let total_pv: f64 = test_prices
                            .iter()
                            .zip(test_volumes.iter())
                            .map(|(p, v)| p * v)
                            .sum();
                        let total_volume: f64 = test_volumes.iter().sum();
                        let scalar_vwap = if total_volume > 0.0 {
                            total_pv / total_volume
                        } else {
                            0.0
                        };

                        black_box((simd_vwap, scalar_vwap));
                    });
                },
            );
        }

        // Test aligned memory performance
        if dispatcher.simd_level() >= SimdLevel::AVX2 {
            let aligned_prices = AlignedPrices::from_slice(test_prices);
            let aligned_volumes = AlignedVolumes::from_slice(test_volumes);

            group.bench_with_input(BenchmarkId::new("vwap_aligned", size), size, |b, _| {
                if let Ok(price_ops) = dispatcher.create_price_ops() {
                    b.iter(|| unsafe {
                        let vwap =
                            price_ops.calculate_vwap_aligned(&aligned_prices, &aligned_volumes);
                        black_box(vwap);
                    });
                }
            });
        }
    }

    group.finish();
}

fn benchmark_lockfree_structures(c: &mut Criterion) {
    let mut group = c.benchmark_group("Lock-Free Structures");

    // Test different buffer sizes
    for size in [256, 1024, 4096].iter() {
        let channel = SharedMemoryChannel::new(*size).expect("Failed to create channel");
        let message = HftMessage::new(message_types::ORDER_REQUEST, [1, 2, 3, 4, 5, 6, 7, 8]);

        group.bench_with_input(BenchmarkId::new("channel_send", size), size, |b, _| {
            b.iter(|| {
                let result = channel.send(message);
                black_box(result);
            });
        });

        // Fill the channel for receive tests
        let _ = channel.send(message);

        group.bench_with_input(BenchmarkId::new("channel_receive", size), size, |b, _| {
            b.iter(|| {
                let result = channel.try_receive();
                black_box(result);
                // Refill for next iteration
                let _ = channel.send(message);
            });
        });

        group.bench_with_input(
            BenchmarkId::new("round_trip_latency", size),
            size,
            |b, _| {
                b.iter(|| {
                    let start = Instant::now();
                    let _ = channel.send(message);
                    let _ = channel.try_receive();
                    let elapsed = start.elapsed();
                    black_box(elapsed);
                });
            },
        );
    }

    group.finish();
}

fn benchmark_end_to_end_latency(c: &mut Criterion) {
    let mut group = c.benchmark_group("End-to-End Latency");

    // Simulate a complete trading operation pipeline
    group.bench_function("complete_trading_pipeline", |b| {
        let dispatcher = SafeSimdDispatcher::new();
        let channel = SharedMemoryChannel::new(1024).expect("Failed to create channel");
        let message = HftMessage::new(message_types::ORDER_REQUEST, [1, 2, 3, 4, 5, 6, 7, 8]);

        // Sample market data
        let prices = vec![100.0, 100.1, 99.9, 100.2];
        let volumes = vec![1000.0, 1500.0, 800.0, 2000.0];

        b.iter(|| {
            let mut total_latency = LatencyMeasurement::start();

            // 1. Market data processing (SIMD VWAP calculation)
            let adaptive_ops = dispatcher.create_adaptive_price_ops();
            let _vwap = adaptive_ops.calculate_vwap(&prices, &volumes);

            // 2. Order validation and risk check (simulated)
            let validation_start = HardwareTimestamp::now();
            std::thread::sleep(Duration::from_nanos(500)); // Simulate validation
            let validation_end = HardwareTimestamp::now();
            let _validation_latency = validation_end.latency_ns(&validation_start);

            // 3. Order routing through lock-free channel
            let _ = channel.send(message);
            let _received = channel.try_receive();

            // 4. Execution response (simulated)
            let execution_start = HardwareTimestamp::now();
            std::thread::sleep(Duration::from_nanos(1000)); // Simulate execution
            let execution_end = HardwareTimestamp::now();
            let _execution_latency = execution_end.latency_ns(&execution_start);

            let total_time = total_latency.finish();
            black_box(total_time);
        });
    });

    group.finish();
}

fn validate_performance_claims(c: &mut Criterion) {
    let mut group = c.benchmark_group("Performance Claims Validation");

    // Validate 14ns RDTSC claim
    group.bench_function("rdtsc_14ns_validation", |b| {
        // Calibrate TSC first
        let _tsc_freq = calibrate_tsc();

        b.iter(|| {
            let start = Instant::now();
            let _ts = HardwareTimestamp::now();
            let elapsed = start.elapsed();
            black_box(elapsed);
        });
    });

    // Validate sub-50μs end-to-end claim
    group.bench_function("sub_50us_validation", |b| {
        let dispatcher = SafeSimdDispatcher::new();
        let channel = SharedMemoryChannel::new(1024).expect("Failed to create channel");
        let prices = vec![100.0; 100];
        let volumes = vec![1000.0; 100];

        b.iter(|| {
            let start = Instant::now();

            // Complete HFT pipeline
            let adaptive_ops = dispatcher.create_adaptive_price_ops();
            let _vwap = adaptive_ops.calculate_vwap(&prices, &volumes);

            let message = HftMessage::new(message_types::ORDER_REQUEST, [1, 2, 3, 4, 5, 6, 7, 8]);
            let _ = channel.send(message);
            let _ = channel.try_receive();

            let elapsed = start.elapsed();
            black_box(elapsed);
        });
    });

    group.finish();
}

criterion_group!(
    benches,
    benchmark_rdtsc_timing,
    benchmark_simd_operations,
    benchmark_lockfree_structures,
    benchmark_end_to_end_latency,
    validate_performance_claims
);
criterion_main!(benches);