Compare commits

...

5 Commits

Author SHA1 Message Date
Alex P 8fb0b9f9c6 feat(audio): centralize audio configuration and improve debugging
- Add debug utilities with development-only logging
- Create centralized audio configuration constants
- Implement audio quality service for managing presets
- Replace console logging with debug utilities
- Update audio metrics with unified structure
- Improve microphone error handling and state management
2025-08-27 13:01:56 +00:00
Alex P e8d12bae4b style(audio): fix formatting and add missing newlines
- Fix indentation in test files and supervisor code
- Add missing newlines at end of files
- Clean up documentation formatting
- Fix buffer pool pointer return type
2025-08-26 16:49:41 +00:00
Alex P 6a68e23d12 refactor(audio): improve error handling and memory management
- remove redundant error logging in audio supervisor stop calls
- add buffer pool for memory optimization in audio relay and ipc
- return default metrics when process is not running
- add channel closed flags to prevent double closing
- standardize component naming and logging
- add comprehensive documentation for audio components
- improve test coverage with new unit tests
2025-08-26 14:36:07 +00:00
Alex P b1f85db7de feat(audio): enhance error handling and add device health monitoring
- Implement robust error recovery with progressive backoff in audio streaming
- Add comprehensive device health monitoring system
- Improve ALSA device handling with enhanced retry logic
- Refactor IPC message handling to use shared pools
- Add validation utilities for audio frames and configuration
- Introduce atomic utilities for thread-safe metrics tracking
- Update latency histogram to use configurable buckets
- Add documentation for new metrics and configuration options
2025-08-26 12:51:11 +00:00
Alex P e4ed2b8fad refactor(audio): rename audio components for clarity and add validation
Rename audio server/client components to be more specific (AudioOutputServer/Client). Add new validation.go and ipc_common.go files for shared IPC functionality. Improve error handling and cleanup in input/output IPC components.

Disable granular metrics logging to reduce log pollution. Reset metrics on failed start and ensure proper cleanup. Add common IPC message interface and optimized message pool for reuse.
2025-08-26 10:42:25 +00:00
46 changed files with 4830 additions and 620 deletions

View File

@ -11,7 +11,27 @@ import (
"github.com/rs/zerolog"
)
// AdaptiveBufferConfig holds configuration for adaptive buffer sizing
// AdaptiveBufferConfig holds configuration for the adaptive buffer sizing algorithm.
//
// The adaptive buffer system dynamically adjusts audio buffer sizes based on real-time
// system conditions to optimize the trade-off between latency and stability. The algorithm
// uses multiple factors to make decisions:
//
// 1. System Load Monitoring:
// - CPU usage: High CPU load increases buffer sizes to prevent underruns
// - Memory usage: High memory pressure reduces buffer sizes to conserve RAM
//
// 2. Latency Tracking:
// - Target latency: Optimal latency for the current quality setting
// - Max latency: Hard limit beyond which buffers are aggressively reduced
//
// 3. Adaptation Strategy:
// - Exponential smoothing: Prevents oscillation and provides stable adjustments
// - Discrete steps: Buffer sizes change in fixed increments to avoid instability
// - Hysteresis: Different thresholds for increasing vs decreasing buffer sizes
//
// The algorithm is specifically tuned for embedded ARM systems with limited resources,
// prioritizing stability over absolute minimum latency.
type AdaptiveBufferConfig struct {
// Buffer size limits (in frames)
MinBufferSize int
@ -156,6 +176,32 @@ func (abm *AdaptiveBufferManager) adaptationLoop() {
}
// adaptBufferSizes analyzes system conditions and adjusts buffer sizes
// adaptBufferSizes implements the core adaptive buffer sizing algorithm.
//
// This function uses a multi-factor approach to determine optimal buffer sizes:
//
// Mathematical Model:
// 1. Factor Calculation:
//
// - CPU Factor: Sigmoid function that increases buffer size under high CPU load
//
// - Memory Factor: Inverse relationship that decreases buffer size under memory pressure
//
// - Latency Factor: Exponential decay that aggressively reduces buffers when latency exceeds targets
//
// 2. Combined Factor:
// Combined = (CPU_factor * Memory_factor * Latency_factor)
// This multiplicative approach ensures any single critical factor can override others
//
// 3. Exponential Smoothing:
// New_size = Current_size + smoothing_factor * (Target_size - Current_size)
// This prevents rapid oscillations and provides stable convergence
//
// 4. Discrete Quantization:
// Final sizes are rounded to frame boundaries and clamped to configured limits
//
// The algorithm runs periodically and only applies changes when the adaptation interval
// has elapsed, preventing excessive adjustments that could destabilize the audio pipeline.
func (abm *AdaptiveBufferManager) adaptBufferSizes() {
// Collect current system metrics
metrics := abm.processMonitor.GetCurrentMetrics()

View File

@ -45,7 +45,7 @@ func DefaultOptimizerConfig() OptimizerConfig {
CooldownPeriod: GetConfig().CooldownPeriod,
Aggressiveness: GetConfig().OptimizerAggressiveness,
RollbackThreshold: GetConfig().RollbackThreshold,
StabilityPeriod: 10 * time.Second,
StabilityPeriod: GetConfig().AdaptiveOptimizerStability,
}
}

View File

@ -9,7 +9,7 @@ import (
var (
// Global audio output supervisor instance
globalOutputSupervisor unsafe.Pointer // *AudioServerSupervisor
globalOutputSupervisor unsafe.Pointer // *AudioOutputSupervisor
)
// isAudioServerProcess detects if we're running as the audio server subprocess
@ -58,15 +58,15 @@ func StopNonBlockingAudioStreaming() {
}
// SetAudioOutputSupervisor sets the global audio output supervisor
func SetAudioOutputSupervisor(supervisor *AudioServerSupervisor) {
func SetAudioOutputSupervisor(supervisor *AudioOutputSupervisor) {
atomic.StorePointer(&globalOutputSupervisor, unsafe.Pointer(supervisor))
}
// GetAudioOutputSupervisor returns the global audio output supervisor
func GetAudioOutputSupervisor() *AudioServerSupervisor {
func GetAudioOutputSupervisor() *AudioOutputSupervisor {
ptr := atomic.LoadPointer(&globalOutputSupervisor)
if ptr == nil {
return nil
}
return (*AudioServerSupervisor)(ptr)
return (*AudioOutputSupervisor)(ptr)
}

View File

@ -0,0 +1,204 @@
package audio
import (
"sync/atomic"
"time"
)
// AtomicCounter provides thread-safe counter operations
type AtomicCounter struct {
value int64
}
// NewAtomicCounter creates a new atomic counter
func NewAtomicCounter() *AtomicCounter {
return &AtomicCounter{}
}
// Add atomically adds delta to the counter and returns the new value
func (c *AtomicCounter) Add(delta int64) int64 {
return atomic.AddInt64(&c.value, delta)
}
// Increment atomically increments the counter by 1
func (c *AtomicCounter) Increment() int64 {
return atomic.AddInt64(&c.value, 1)
}
// Load atomically loads the counter value
func (c *AtomicCounter) Load() int64 {
return atomic.LoadInt64(&c.value)
}
// Store atomically stores a new value
func (c *AtomicCounter) Store(value int64) {
atomic.StoreInt64(&c.value, value)
}
// Reset atomically resets the counter to zero
func (c *AtomicCounter) Reset() {
atomic.StoreInt64(&c.value, 0)
}
// Swap atomically swaps the value and returns the old value
func (c *AtomicCounter) Swap(new int64) int64 {
return atomic.SwapInt64(&c.value, new)
}
// FrameMetrics provides common frame tracking metrics
type FrameMetrics struct {
Total *AtomicCounter
Dropped *AtomicCounter
Bytes *AtomicCounter
}
// NewFrameMetrics creates a new frame metrics tracker
func NewFrameMetrics() *FrameMetrics {
return &FrameMetrics{
Total: NewAtomicCounter(),
Dropped: NewAtomicCounter(),
Bytes: NewAtomicCounter(),
}
}
// RecordFrame atomically records a successful frame with its size
func (fm *FrameMetrics) RecordFrame(size int64) {
fm.Total.Increment()
fm.Bytes.Add(size)
}
// RecordDrop atomically records a dropped frame
func (fm *FrameMetrics) RecordDrop() {
fm.Dropped.Increment()
}
// GetStats returns current metrics values
func (fm *FrameMetrics) GetStats() (total, dropped, bytes int64) {
return fm.Total.Load(), fm.Dropped.Load(), fm.Bytes.Load()
}
// Reset resets all metrics to zero
func (fm *FrameMetrics) Reset() {
fm.Total.Reset()
fm.Dropped.Reset()
fm.Bytes.Reset()
}
// GetDropRate calculates the drop rate as a percentage
func (fm *FrameMetrics) GetDropRate() float64 {
total := fm.Total.Load()
if total == 0 {
return 0.0
}
dropped := fm.Dropped.Load()
return float64(dropped) / float64(total) * 100.0
}
// LatencyTracker provides atomic latency tracking
type LatencyTracker struct {
current *AtomicCounter
min *AtomicCounter
max *AtomicCounter
average *AtomicCounter
samples *AtomicCounter
}
// NewLatencyTracker creates a new latency tracker
func NewLatencyTracker() *LatencyTracker {
lt := &LatencyTracker{
current: NewAtomicCounter(),
min: NewAtomicCounter(),
max: NewAtomicCounter(),
average: NewAtomicCounter(),
samples: NewAtomicCounter(),
}
// Initialize min to max value so first measurement sets it properly
lt.min.Store(int64(^uint64(0) >> 1)) // Max int64
return lt
}
// RecordLatency atomically records a new latency measurement
func (lt *LatencyTracker) RecordLatency(latency time.Duration) {
latencyNanos := latency.Nanoseconds()
lt.current.Store(latencyNanos)
lt.samples.Increment()
// Update min
for {
oldMin := lt.min.Load()
if latencyNanos >= oldMin {
break
}
if atomic.CompareAndSwapInt64(&lt.min.value, oldMin, latencyNanos) {
break
}
}
// Update max
for {
oldMax := lt.max.Load()
if latencyNanos <= oldMax {
break
}
if atomic.CompareAndSwapInt64(&lt.max.value, oldMax, latencyNanos) {
break
}
}
// Update average using exponential moving average
oldAvg := lt.average.Load()
newAvg := (oldAvg*7 + latencyNanos) / 8 // 87.5% weight to old average
lt.average.Store(newAvg)
}
// GetLatencyStats returns current latency statistics
func (lt *LatencyTracker) GetLatencyStats() (current, min, max, average time.Duration, samples int64) {
return time.Duration(lt.current.Load()),
time.Duration(lt.min.Load()),
time.Duration(lt.max.Load()),
time.Duration(lt.average.Load()),
lt.samples.Load()
}
// PoolMetrics provides common pool performance metrics
type PoolMetrics struct {
Hits *AtomicCounter
Misses *AtomicCounter
}
// NewPoolMetrics creates a new pool metrics tracker
func NewPoolMetrics() *PoolMetrics {
return &PoolMetrics{
Hits: NewAtomicCounter(),
Misses: NewAtomicCounter(),
}
}
// RecordHit atomically records a pool hit
func (pm *PoolMetrics) RecordHit() {
pm.Hits.Increment()
}
// RecordMiss atomically records a pool miss
func (pm *PoolMetrics) RecordMiss() {
pm.Misses.Increment()
}
// GetHitRate calculates the hit rate as a percentage
func (pm *PoolMetrics) GetHitRate() float64 {
hits := pm.Hits.Load()
misses := pm.Misses.Load()
total := hits + misses
if total == 0 {
return 0.0
}
return float64(hits) / float64(total) * 100.0
}
// GetStats returns hit and miss counts
func (pm *PoolMetrics) GetStats() (hits, misses int64, hitRate float64) {
hits = pm.Hits.Load()
misses = pm.Misses.Load()
hitRate = pm.GetHitRate()
return
}

View File

@ -40,7 +40,8 @@ func NewAudioBufferPool(bufferSize int) *AudioBufferPool {
preallocSize: preallocSize,
pool: sync.Pool{
New: func() interface{} {
return make([]byte, 0, bufferSize)
buf := make([]byte, 0, bufferSize)
return &buf
},
},
}

View File

@ -61,12 +61,15 @@ static volatile int capture_initialized = 0;
static volatile int playback_initializing = 0;
static volatile int playback_initialized = 0;
// Safe ALSA device opening with retry logic
// Enhanced ALSA device opening with exponential backoff retry logic
static int safe_alsa_open(snd_pcm_t **handle, const char *device, snd_pcm_stream_t stream) {
int attempts = 3;
int max_attempts = 5; // Increased from 3 to 5
int attempt = 0;
int err;
int backoff_us = sleep_microseconds; // Start with base sleep time
const int max_backoff_us = 500000; // Max 500ms backoff
while (attempts-- > 0) {
while (attempt < max_attempts) {
err = snd_pcm_open(handle, device, stream, SND_PCM_NONBLOCK);
if (err >= 0) {
// Switch to blocking mode after successful open
@ -74,12 +77,26 @@ static int safe_alsa_open(snd_pcm_t **handle, const char *device, snd_pcm_stream
return 0;
}
if (err == -EBUSY && attempts > 0) {
// Device busy, wait and retry
usleep(sleep_microseconds); // 50ms
continue;
attempt++;
if (attempt >= max_attempts) break;
// Enhanced error handling with specific retry strategies
if (err == -EBUSY || err == -EAGAIN) {
// Device busy or temporarily unavailable - retry with backoff
usleep(backoff_us);
backoff_us = (backoff_us * 2 < max_backoff_us) ? backoff_us * 2 : max_backoff_us;
} else if (err == -ENODEV || err == -ENOENT) {
// Device not found - longer wait as device might be initializing
usleep(backoff_us * 2);
backoff_us = (backoff_us * 2 < max_backoff_us) ? backoff_us * 2 : max_backoff_us;
} else if (err == -EPERM || err == -EACCES) {
// Permission denied - shorter wait, likely persistent issue
usleep(backoff_us / 2);
} else {
// Other errors - standard backoff
usleep(backoff_us);
backoff_us = (backoff_us * 2 < max_backoff_us) ? backoff_us * 2 : max_backoff_us;
}
break;
}
return err;
}
@ -217,43 +234,114 @@ int jetkvm_audio_init() {
return 0;
}
// Read and encode one frame with enhanced error handling
// jetkvm_audio_read_encode reads one audio frame from ALSA, encodes it with Opus, and handles errors.
//
// This function implements a robust audio capture pipeline with the following features:
// - ALSA PCM capture with automatic device recovery
// - Opus encoding with optimized settings for real-time processing
// - Progressive error recovery with exponential backoff
// - Buffer underrun and device suspension handling
//
// Error Recovery Strategy:
// 1. EPIPE (buffer underrun): Prepare device and retry with progressive delays
// 2. ESTRPIPE (device suspended): Resume device with timeout and fallback to prepare
// 3. Other errors: Log and attempt recovery up to max_recovery_attempts
//
// Performance Optimizations:
// - Stack-allocated PCM buffer to avoid heap allocations
// - Direct memory access for Opus encoding
// - Minimal system calls in the hot path
//
// Parameters:
// opus_buf: Output buffer for encoded Opus data (must be at least max_packet_size bytes)
//
// Returns:
// >0: Number of bytes written to opus_buf
// -1: Initialization error or safety check failure
// -2: Unrecoverable ALSA or Opus error after all retry attempts
int jetkvm_audio_read_encode(void *opus_buf) {
short pcm_buffer[1920]; // max 2ch*960
unsigned char *out = (unsigned char*)opus_buf;
int err = 0;
int recovery_attempts = 0;
const int max_recovery_attempts = 3;
// Safety checks
if (!capture_initialized || !pcm_handle || !encoder || !opus_buf) {
return -1;
}
retry_read:
;
int pcm_rc = snd_pcm_readi(pcm_handle, pcm_buffer, frame_size);
// Handle ALSA errors with enhanced recovery
// Handle ALSA errors with robust recovery strategies
if (pcm_rc < 0) {
if (pcm_rc == -EPIPE) {
// Buffer underrun - try to recover
err = snd_pcm_prepare(pcm_handle);
if (err < 0) return -1;
// Buffer underrun - implement progressive recovery
recovery_attempts++;
if (recovery_attempts > max_recovery_attempts) {
return -1; // Give up after max attempts
}
pcm_rc = snd_pcm_readi(pcm_handle, pcm_buffer, frame_size);
if (pcm_rc < 0) return -1;
// Try to recover with prepare
err = snd_pcm_prepare(pcm_handle);
if (err < 0) {
// If prepare fails, try drop and prepare
snd_pcm_drop(pcm_handle);
err = snd_pcm_prepare(pcm_handle);
if (err < 0) return -1;
}
// Wait before retry to allow device to stabilize
usleep(sleep_microseconds * recovery_attempts);
goto retry_read;
} else if (pcm_rc == -EAGAIN) {
// No data available - return 0 to indicate no frame
return 0;
} else if (pcm_rc == -ESTRPIPE) {
// Device suspended, try to resume
while ((err = snd_pcm_resume(pcm_handle)) == -EAGAIN) {
usleep(sleep_microseconds); // Use centralized constant
// Device suspended, implement robust resume logic
recovery_attempts++;
if (recovery_attempts > max_recovery_attempts) {
return -1;
}
// Try to resume with timeout
int resume_attempts = 0;
while ((err = snd_pcm_resume(pcm_handle)) == -EAGAIN && resume_attempts < 10) {
usleep(sleep_microseconds);
resume_attempts++;
}
if (err < 0) {
// Resume failed, try prepare as fallback
err = snd_pcm_prepare(pcm_handle);
if (err < 0) return -1;
}
return 0; // Skip this frame
// Wait before retry to allow device to stabilize
usleep(sleep_microseconds * recovery_attempts);
return 0; // Skip this frame but don't fail
} else if (pcm_rc == -ENODEV) {
// Device disconnected - critical error
return -1;
} else if (pcm_rc == -EIO) {
// I/O error - try recovery once
recovery_attempts++;
if (recovery_attempts <= max_recovery_attempts) {
snd_pcm_drop(pcm_handle);
err = snd_pcm_prepare(pcm_handle);
if (err >= 0) {
usleep(sleep_microseconds);
goto retry_read;
}
}
return -1;
} else {
// Other error - return error code
// Other errors - limited retry for transient issues
recovery_attempts++;
if (recovery_attempts <= 1 && (pcm_rc == -EINTR || pcm_rc == -EBUSY)) {
usleep(sleep_microseconds / 2);
goto retry_read;
}
return -1;
}
}
@ -327,11 +415,38 @@ int jetkvm_audio_playback_init() {
return 0;
}
// Decode Opus and write PCM with enhanced error handling
// jetkvm_audio_decode_write decodes Opus data and writes PCM to ALSA playback device.
//
// This function implements a robust audio playback pipeline with the following features:
// - Opus decoding with packet loss concealment
// - ALSA PCM playback with automatic device recovery
// - Progressive error recovery with exponential backoff
// - Buffer underrun and device suspension handling
//
// Error Recovery Strategy:
// 1. EPIPE (buffer underrun): Prepare device, optionally drop+prepare, retry with delays
// 2. ESTRPIPE (device suspended): Resume with timeout, fallback to prepare if needed
// 3. Opus decode errors: Attempt packet loss concealment before failing
//
// Performance Optimizations:
// - Stack-allocated PCM buffer to minimize heap allocations
// - Bounds checking to prevent buffer overruns
// - Direct ALSA device access for minimal latency
//
// Parameters:
// opus_buf: Input buffer containing Opus-encoded audio data
// opus_size: Size of the Opus data in bytes (must be > 0 and <= max_packet_size)
//
// Returns:
// 0: Success - audio frame decoded and written to playback device
// -1: Invalid parameters, initialization error, or bounds check failure
// -2: Unrecoverable ALSA or Opus error after all retry attempts
int jetkvm_audio_decode_write(void *opus_buf, int opus_size) {
short pcm_buffer[1920]; // max 2ch*960
unsigned char *in = (unsigned char*)opus_buf;
int err = 0;
int recovery_attempts = 0;
const int max_recovery_attempts = 3;
// Safety checks
if (!playback_initialized || !pcm_playback_handle || !decoder || !opus_buf || opus_size <= 0) {
@ -343,31 +458,91 @@ int jetkvm_audio_decode_write(void *opus_buf, int opus_size) {
return -1;
}
// Decode Opus to PCM
// Decode Opus to PCM with error handling
int pcm_frames = opus_decode(decoder, in, opus_size, pcm_buffer, frame_size, 0);
if (pcm_frames < 0) return -1;
if (pcm_frames < 0) {
// Try packet loss concealment on decode error
pcm_frames = opus_decode(decoder, NULL, 0, pcm_buffer, frame_size, 0);
if (pcm_frames < 0) return -1;
}
// Write PCM to playback device with enhanced recovery
retry_write:
;
// Write PCM to playback device with robust recovery
int pcm_rc = snd_pcm_writei(pcm_playback_handle, pcm_buffer, pcm_frames);
if (pcm_rc < 0) {
if (pcm_rc == -EPIPE) {
// Buffer underrun - try to recover
err = snd_pcm_prepare(pcm_playback_handle);
if (err < 0) return -2;
pcm_rc = snd_pcm_writei(pcm_playback_handle, pcm_buffer, pcm_frames);
} else if (pcm_rc == -ESTRPIPE) {
// Device suspended, try to resume
while ((err = snd_pcm_resume(pcm_playback_handle)) == -EAGAIN) {
usleep(sleep_microseconds); // Use centralized constant
// Buffer underrun - implement progressive recovery
recovery_attempts++;
if (recovery_attempts > max_recovery_attempts) {
return -2;
}
// Try to recover with prepare
err = snd_pcm_prepare(pcm_playback_handle);
if (err < 0) {
// If prepare fails, try drop and prepare
snd_pcm_drop(pcm_playback_handle);
err = snd_pcm_prepare(pcm_playback_handle);
if (err < 0) return -2;
}
return 0; // Skip this frame
// Wait before retry to allow device to stabilize
usleep(sleep_microseconds * recovery_attempts);
goto retry_write;
} else if (pcm_rc == -ESTRPIPE) {
// Device suspended, implement robust resume logic
recovery_attempts++;
if (recovery_attempts > max_recovery_attempts) {
return -2;
}
// Try to resume with timeout
int resume_attempts = 0;
while ((err = snd_pcm_resume(pcm_playback_handle)) == -EAGAIN && resume_attempts < 10) {
usleep(sleep_microseconds);
resume_attempts++;
}
if (err < 0) {
// Resume failed, try prepare as fallback
err = snd_pcm_prepare(pcm_playback_handle);
if (err < 0) return -2;
}
// Wait before retry to allow device to stabilize
usleep(sleep_microseconds * recovery_attempts);
return 0; // Skip this frame but don't fail
} else if (pcm_rc == -ENODEV) {
// Device disconnected - critical error
return -2;
} else if (pcm_rc == -EIO) {
// I/O error - try recovery once
recovery_attempts++;
if (recovery_attempts <= max_recovery_attempts) {
snd_pcm_drop(pcm_playback_handle);
err = snd_pcm_prepare(pcm_playback_handle);
if (err >= 0) {
usleep(sleep_microseconds);
goto retry_write;
}
}
return -2;
} else if (pcm_rc == -EAGAIN) {
// Device not ready - brief wait and retry
recovery_attempts++;
if (recovery_attempts <= max_recovery_attempts) {
usleep(sleep_microseconds / 4);
goto retry_write;
}
return -2;
} else {
// Other errors - limited retry for transient issues
recovery_attempts++;
if (recovery_attempts <= 1 && (pcm_rc == -EINTR || pcm_rc == -EBUSY)) {
usleep(sleep_microseconds / 2);
goto retry_write;
}
return -2;
}
if (pcm_rc < 0) return -2;
}
return pcm_frames;

View File

@ -881,6 +881,12 @@ type AudioConfigConstants struct {
// Default 5s provides responsive input monitoring.
InputSupervisorTimeout time.Duration // 5s
// OutputSupervisorTimeout defines timeout for output supervisor operations.
// Used in: supervisor.go for output process monitoring
// Impact: Shorter timeouts improve output responsiveness but may cause false timeouts.
// Default 5s provides responsive output monitoring.
OutputSupervisorTimeout time.Duration // 5s
// ShortTimeout defines brief timeout for time-critical operations.
// Used in: Real-time audio processing for minimal timeout scenarios
// Impact: Very short timeouts ensure responsiveness but may cause premature failures.
@ -1382,6 +1388,201 @@ type AudioConfigConstants struct {
// Impact: Controls scaling factor for memory influence on buffer sizing.
// Default 100 provides standard percentage scaling for memory calculations.
AdaptiveBufferMemoryMultiplier int
// Socket Names - Configuration for IPC socket file names
// Used in: IPC communication for audio input/output
// Impact: Controls socket file naming and IPC connection endpoints
// InputSocketName defines the socket file name for audio input IPC.
// Used in: input_ipc.go for microphone input communication
// Impact: Must be unique to prevent conflicts with other audio sockets.
// Default "audio_input.sock" provides clear identification for input socket.
InputSocketName string
// OutputSocketName defines the socket file name for audio output IPC.
// Used in: ipc.go for audio output communication
// Impact: Must be unique to prevent conflicts with other audio sockets.
// Default "audio_output.sock" provides clear identification for output socket.
OutputSocketName string
// Component Names - Standardized component identifiers for logging
// Used in: Logging and monitoring throughout audio system
// Impact: Provides consistent component identification across logs
// AudioInputComponentName defines component name for audio input logging.
// Used in: input_ipc.go and related input processing components
// Impact: Ensures consistent logging identification for input components.
// Default "audio-input" provides clear component identification.
AudioInputComponentName string
// AudioOutputComponentName defines component name for audio output logging.
// Used in: ipc.go and related output processing components
// Impact: Ensures consistent logging identification for output components.
// Default "audio-output" provides clear component identification.
AudioOutputComponentName string
// AudioServerComponentName defines component name for audio server logging.
// Used in: supervisor.go and server management components
// Impact: Ensures consistent logging identification for server components.
// Default "audio-server" provides clear component identification.
AudioServerComponentName string
// AudioRelayComponentName defines component name for audio relay logging.
// Used in: relay.go for audio relay operations
// Impact: Ensures consistent logging identification for relay components.
// Default "audio-relay" provides clear component identification.
AudioRelayComponentName string
// AudioEventsComponentName defines component name for audio events logging.
// Used in: events.go for event broadcasting operations
// Impact: Ensures consistent logging identification for event components.
// Default "audio-events" provides clear component identification.
AudioEventsComponentName string
// Test Configuration - Constants for testing scenarios
// Used in: Test files for consistent test configuration
// Impact: Provides standardized test parameters and timeouts
// TestSocketTimeout defines timeout for test socket operations.
// Used in: integration_test.go for test socket communication
// Impact: Prevents test hangs while allowing sufficient time for operations.
// Default 100ms provides quick test execution with adequate timeout.
TestSocketTimeout time.Duration
// TestBufferSize defines buffer size for test operations.
// Used in: test_utils.go for test buffer allocation
// Impact: Provides adequate buffer space for test scenarios.
// Default 4096 bytes matches production buffer sizes for realistic testing.
TestBufferSize int
// TestRetryDelay defines delay between test retry attempts.
// Used in: Test files for retry logic in test scenarios
// Impact: Provides reasonable delay for test retry operations.
// Default 200ms allows sufficient time for test state changes.
TestRetryDelay time.Duration
// Latency Histogram Configuration - Constants for latency tracking
// Used in: granular_metrics.go for latency distribution analysis
// Impact: Controls granularity and accuracy of latency measurements
// LatencyHistogramMaxSamples defines maximum samples for latency tracking.
// Used in: granular_metrics.go for latency histogram management
// Impact: Controls memory usage and accuracy of latency statistics.
// Default 1000 samples provides good statistical accuracy with reasonable memory usage.
LatencyHistogramMaxSamples int
// LatencyPercentile50 defines 50th percentile calculation factor.
// Used in: granular_metrics.go for median latency calculation
// Impact: Must be 50 for accurate median calculation.
// Default 50 provides standard median percentile calculation.
LatencyPercentile50 int
// LatencyPercentile95 defines 95th percentile calculation factor.
// Used in: granular_metrics.go for high-percentile latency calculation
// Impact: Must be 95 for accurate 95th percentile calculation.
// Default 95 provides standard high-percentile calculation.
LatencyPercentile95 int
// LatencyPercentile99 defines 99th percentile calculation factor.
// Used in: granular_metrics.go for extreme latency calculation
// Impact: Must be 99 for accurate 99th percentile calculation.
// Default 99 provides standard extreme percentile calculation.
LatencyPercentile99 int
// BufferPoolMaxOperations defines maximum operations to track for efficiency.
// Used in: granular_metrics.go for buffer pool efficiency tracking
// Impact: Controls memory usage and accuracy of efficiency statistics.
// Default 1000 operations provides good balance of accuracy and memory usage.
BufferPoolMaxOperations int
// HitRateCalculationBase defines base value for hit rate percentage calculation.
// Used in: granular_metrics.go for hit rate percentage calculation
// Impact: Must be 100 for accurate percentage calculation.
// Default 100 provides standard percentage calculation base.
HitRateCalculationBase float64
// Validation Constants - Configuration for input validation
// Used in: validation.go for parameter validation
// Impact: Controls validation thresholds and limits
// MaxLatency defines maximum allowed latency for audio processing.
// Used in: validation.go for latency validation
// Impact: Controls maximum acceptable latency before optimization triggers.
// Default 200ms provides reasonable upper bound for real-time audio.
MaxLatency time.Duration
// MinMetricsUpdateInterval defines minimum allowed metrics update interval.
// Used in: validation.go for metrics interval validation
// Impact: Prevents excessive metrics updates that could impact performance.
// Default 100ms provides reasonable minimum update frequency.
MinMetricsUpdateInterval time.Duration
// MaxMetricsUpdateInterval defines maximum allowed metrics update interval.
// Used in: validation.go for metrics interval validation
// Impact: Ensures metrics are updated frequently enough for monitoring.
// Default 30s provides reasonable maximum update interval.
MaxMetricsUpdateInterval time.Duration
// MinSampleRate defines minimum allowed audio sample rate.
// Used in: validation.go for sample rate validation
// Impact: Ensures sample rate is sufficient for audio quality.
// Default 8000Hz provides minimum for voice communication.
MinSampleRate int
// MaxSampleRate defines maximum allowed audio sample rate.
// Used in: validation.go for sample rate validation
// Impact: Prevents excessive sample rates that could impact performance.
// Default 192000Hz provides upper bound for high-quality audio.
MaxSampleRate int
// MaxChannels defines maximum allowed audio channels.
// Used in: validation.go for channel count validation
// Impact: Prevents excessive channel counts that could impact performance.
// Default 8 channels provides reasonable upper bound for multi-channel audio.
MaxChannels int
// Device Health Monitoring Configuration
// Used in: device_health.go for proactive device monitoring and recovery
// Impact: Controls health check frequency and recovery thresholds
// HealthCheckIntervalMS defines interval between device health checks in milliseconds.
// Used in: DeviceHealthMonitor for periodic health assessment
// Impact: Lower values provide faster detection but increase CPU usage.
// Default 5000ms (5s) provides good balance between responsiveness and overhead.
HealthCheckIntervalMS int
// HealthRecoveryThreshold defines number of consecutive successful operations
// required to mark a device as healthy after being unhealthy.
// Used in: DeviceHealthMonitor for recovery state management
// Impact: Higher values prevent premature recovery declarations.
// Default 3 consecutive successes ensures stable recovery.
HealthRecoveryThreshold int
// HealthLatencyThresholdMS defines maximum acceptable latency in milliseconds
// before considering a device unhealthy.
// Used in: DeviceHealthMonitor for latency-based health assessment
// Impact: Lower values trigger recovery sooner but may cause false positives.
// Default 100ms provides reasonable threshold for real-time audio.
HealthLatencyThresholdMS int
// HealthErrorRateLimit defines maximum error rate (0.0-1.0) before
// considering a device unhealthy.
// Used in: DeviceHealthMonitor for error rate assessment
// Impact: Lower values trigger recovery sooner for error-prone devices.
// Default 0.1 (10%) allows some transient errors while detecting problems.
HealthErrorRateLimit float64
// Latency Histogram Bucket Configuration
// Used in: LatencyHistogram for granular latency measurement buckets
// Impact: Defines the boundaries for latency distribution analysis
LatencyBucket10ms time.Duration // 10ms latency bucket
LatencyBucket25ms time.Duration // 25ms latency bucket
LatencyBucket50ms time.Duration // 50ms latency bucket
LatencyBucket100ms time.Duration // 100ms latency bucket
LatencyBucket250ms time.Duration // 250ms latency bucket
LatencyBucket500ms time.Duration // 500ms latency bucket
LatencyBucket1s time.Duration // 1s latency bucket
LatencyBucket2s time.Duration // 2s latency bucket
}
// DefaultAudioConfig returns the default configuration constants
@ -2204,6 +2405,12 @@ func DefaultAudioConfig() *AudioConfigConstants {
// Default 5s (shorter than general supervisor) for faster input recovery
InputSupervisorTimeout: 5 * time.Second,
// OutputSupervisorTimeout defines timeout for output supervisor operations.
// Used in: Output process monitoring, speaker supervision
// Impact: Controls responsiveness of output failure detection
// Default 5s (shorter than general supervisor) for faster output recovery
OutputSupervisorTimeout: 5 * time.Second,
// ShortTimeout defines brief timeout for quick operations (5ms).
// Used in: Lock acquisition, quick IPC operations, immediate responses
// Impact: Critical for maintaining real-time performance
@ -2365,6 +2572,56 @@ func DefaultAudioConfig() *AudioConfigConstants {
// Adaptive Buffer Constants
AdaptiveBufferCPUMultiplier: 100, // 100 multiplier for CPU percentage
AdaptiveBufferMemoryMultiplier: 100, // 100 multiplier for memory percentage
// Socket Names
InputSocketName: "audio_input.sock", // Socket name for audio input IPC
OutputSocketName: "audio_output.sock", // Socket name for audio output IPC
// Component Names
AudioInputComponentName: "audio-input", // Component name for input logging
AudioOutputComponentName: "audio-output", // Component name for output logging
AudioServerComponentName: "audio-server", // Component name for server logging
AudioRelayComponentName: "audio-relay", // Component name for relay logging
AudioEventsComponentName: "audio-events", // Component name for events logging
// Test Configuration
TestSocketTimeout: 100 * time.Millisecond, // 100ms timeout for test socket operations
TestBufferSize: 4096, // 4096 bytes buffer size for test operations
TestRetryDelay: 200 * time.Millisecond, // 200ms delay between test retry attempts
// Latency Histogram Configuration
LatencyHistogramMaxSamples: 1000, // 1000 samples for latency tracking
LatencyPercentile50: 50, // 50th percentile calculation factor
LatencyPercentile95: 95, // 95th percentile calculation factor
LatencyPercentile99: 99, // 99th percentile calculation factor
// Buffer Pool Efficiency Constants
BufferPoolMaxOperations: 1000, // 1000 operations for efficiency tracking
HitRateCalculationBase: 100.0, // 100.0 base for hit rate percentage calculation
// Validation Constants
MaxLatency: 500 * time.Millisecond, // 500ms maximum allowed latency
MinMetricsUpdateInterval: 100 * time.Millisecond, // 100ms minimum metrics update interval
MaxMetricsUpdateInterval: 10 * time.Second, // 10s maximum metrics update interval
MinSampleRate: 8000, // 8kHz minimum sample rate
MaxSampleRate: 48000, // 48kHz maximum sample rate
MaxChannels: 8, // 8 maximum audio channels
// Device Health Monitoring Configuration
HealthCheckIntervalMS: 5000, // 5000ms (5s) health check interval
HealthRecoveryThreshold: 3, // 3 consecutive successes for recovery
HealthLatencyThresholdMS: 100, // 100ms latency threshold for health
HealthErrorRateLimit: 0.1, // 10% error rate limit for health
// Latency Histogram Bucket Configuration
LatencyBucket10ms: 10 * time.Millisecond, // 10ms latency bucket
LatencyBucket25ms: 25 * time.Millisecond, // 25ms latency bucket
LatencyBucket50ms: 50 * time.Millisecond, // 50ms latency bucket
LatencyBucket100ms: 100 * time.Millisecond, // 100ms latency bucket
LatencyBucket250ms: 250 * time.Millisecond, // 250ms latency bucket
LatencyBucket500ms: 500 * time.Millisecond, // 500ms latency bucket
LatencyBucket1s: 1 * time.Second, // 1s latency bucket
LatencyBucket2s: 2 * time.Second, // 2s latency bucket
}
}

View File

@ -0,0 +1,514 @@
package audio
import (
"context"
"fmt"
"sync"
"sync/atomic"
"time"
"github.com/jetkvm/kvm/internal/logging"
"github.com/rs/zerolog"
)
// DeviceHealthStatus represents the health status of an audio device
type DeviceHealthStatus int
const (
DeviceHealthUnknown DeviceHealthStatus = iota
DeviceHealthHealthy
DeviceHealthDegraded
DeviceHealthFailing
DeviceHealthCritical
)
func (s DeviceHealthStatus) String() string {
switch s {
case DeviceHealthHealthy:
return "healthy"
case DeviceHealthDegraded:
return "degraded"
case DeviceHealthFailing:
return "failing"
case DeviceHealthCritical:
return "critical"
default:
return "unknown"
}
}
// DeviceHealthMetrics tracks health-related metrics for audio devices
type DeviceHealthMetrics struct {
// Error tracking
ConsecutiveErrors int64 `json:"consecutive_errors"`
TotalErrors int64 `json:"total_errors"`
LastErrorTime time.Time `json:"last_error_time"`
ErrorRate float64 `json:"error_rate"` // errors per minute
// Performance metrics
AverageLatency time.Duration `json:"average_latency"`
MaxLatency time.Duration `json:"max_latency"`
LatencySpikes int64 `json:"latency_spikes"`
Underruns int64 `json:"underruns"`
Overruns int64 `json:"overruns"`
// Device availability
LastSuccessfulOp time.Time `json:"last_successful_op"`
DeviceDisconnects int64 `json:"device_disconnects"`
RecoveryAttempts int64 `json:"recovery_attempts"`
SuccessfulRecoveries int64 `json:"successful_recoveries"`
// Health assessment
CurrentStatus DeviceHealthStatus `json:"current_status"`
StatusLastChanged time.Time `json:"status_last_changed"`
HealthScore float64 `json:"health_score"` // 0.0 to 1.0
}
// DeviceHealthMonitor monitors the health of audio devices and triggers recovery
type DeviceHealthMonitor struct {
// Atomic fields first for ARM32 alignment
running int32
monitoringEnabled int32
// Configuration
checkInterval time.Duration
recoveryThreshold int
latencyThreshold time.Duration
errorRateLimit float64 // max errors per minute
// State tracking
captureMetrics *DeviceHealthMetrics
playbackMetrics *DeviceHealthMetrics
mutex sync.RWMutex
// Control channels
ctx context.Context
cancel context.CancelFunc
stopChan chan struct{}
doneChan chan struct{}
// Recovery callbacks
recoveryCallbacks map[string]func() error
callbackMutex sync.RWMutex
// Logging
logger zerolog.Logger
config *AudioConfigConstants
}
// NewDeviceHealthMonitor creates a new device health monitor
func NewDeviceHealthMonitor() *DeviceHealthMonitor {
ctx, cancel := context.WithCancel(context.Background())
config := GetConfig()
return &DeviceHealthMonitor{
checkInterval: time.Duration(config.HealthCheckIntervalMS) * time.Millisecond,
recoveryThreshold: config.HealthRecoveryThreshold,
latencyThreshold: time.Duration(config.HealthLatencyThresholdMS) * time.Millisecond,
errorRateLimit: config.HealthErrorRateLimit,
captureMetrics: &DeviceHealthMetrics{
CurrentStatus: DeviceHealthUnknown,
HealthScore: 1.0,
},
playbackMetrics: &DeviceHealthMetrics{
CurrentStatus: DeviceHealthUnknown,
HealthScore: 1.0,
},
ctx: ctx,
cancel: cancel,
stopChan: make(chan struct{}),
doneChan: make(chan struct{}),
recoveryCallbacks: make(map[string]func() error),
logger: logging.GetDefaultLogger().With().Str("component", "device-health-monitor").Logger(),
config: config,
}
}
// Start begins health monitoring
func (dhm *DeviceHealthMonitor) Start() error {
if !atomic.CompareAndSwapInt32(&dhm.running, 0, 1) {
return fmt.Errorf("device health monitor already running")
}
dhm.logger.Info().Msg("starting device health monitor")
atomic.StoreInt32(&dhm.monitoringEnabled, 1)
go dhm.monitoringLoop()
return nil
}
// Stop stops health monitoring
func (dhm *DeviceHealthMonitor) Stop() {
if !atomic.CompareAndSwapInt32(&dhm.running, 1, 0) {
return
}
dhm.logger.Info().Msg("stopping device health monitor")
atomic.StoreInt32(&dhm.monitoringEnabled, 0)
close(dhm.stopChan)
dhm.cancel()
// Wait for monitoring loop to finish
select {
case <-dhm.doneChan:
dhm.logger.Info().Msg("device health monitor stopped")
case <-time.After(time.Duration(dhm.config.SupervisorTimeout)):
dhm.logger.Warn().Msg("device health monitor stop timeout")
}
}
// RegisterRecoveryCallback registers a recovery function for a specific component
func (dhm *DeviceHealthMonitor) RegisterRecoveryCallback(component string, callback func() error) {
dhm.callbackMutex.Lock()
defer dhm.callbackMutex.Unlock()
dhm.recoveryCallbacks[component] = callback
dhm.logger.Info().Str("component", component).Msg("registered recovery callback")
}
// RecordError records an error for health tracking
func (dhm *DeviceHealthMonitor) RecordError(deviceType string, err error) {
if atomic.LoadInt32(&dhm.monitoringEnabled) == 0 {
return
}
dhm.mutex.Lock()
defer dhm.mutex.Unlock()
var metrics *DeviceHealthMetrics
switch deviceType {
case "capture":
metrics = dhm.captureMetrics
case "playback":
metrics = dhm.playbackMetrics
default:
dhm.logger.Warn().Str("device_type", deviceType).Msg("unknown device type for error recording")
return
}
atomic.AddInt64(&metrics.ConsecutiveErrors, 1)
atomic.AddInt64(&metrics.TotalErrors, 1)
metrics.LastErrorTime = time.Now()
// Update error rate (errors per minute)
if !metrics.LastErrorTime.IsZero() {
timeSinceFirst := time.Since(metrics.LastErrorTime)
if timeSinceFirst > 0 {
metrics.ErrorRate = float64(metrics.TotalErrors) / timeSinceFirst.Minutes()
}
}
dhm.logger.Debug().
Str("device_type", deviceType).
Err(err).
Int64("consecutive_errors", metrics.ConsecutiveErrors).
Float64("error_rate", metrics.ErrorRate).
Msg("recorded device error")
// Trigger immediate health assessment
dhm.assessDeviceHealth(deviceType, metrics)
}
// RecordSuccess records a successful operation
func (dhm *DeviceHealthMonitor) RecordSuccess(deviceType string) {
if atomic.LoadInt32(&dhm.monitoringEnabled) == 0 {
return
}
dhm.mutex.Lock()
defer dhm.mutex.Unlock()
var metrics *DeviceHealthMetrics
switch deviceType {
case "capture":
metrics = dhm.captureMetrics
case "playback":
metrics = dhm.playbackMetrics
default:
return
}
// Reset consecutive errors on success
atomic.StoreInt64(&metrics.ConsecutiveErrors, 0)
metrics.LastSuccessfulOp = time.Now()
// Improve health score gradually
if metrics.HealthScore < 1.0 {
metrics.HealthScore = min(1.0, metrics.HealthScore+0.1)
}
}
// RecordLatency records operation latency for health assessment
func (dhm *DeviceHealthMonitor) RecordLatency(deviceType string, latency time.Duration) {
if atomic.LoadInt32(&dhm.monitoringEnabled) == 0 {
return
}
dhm.mutex.Lock()
defer dhm.mutex.Unlock()
var metrics *DeviceHealthMetrics
switch deviceType {
case "capture":
metrics = dhm.captureMetrics
case "playback":
metrics = dhm.playbackMetrics
default:
return
}
// Update latency metrics
if metrics.AverageLatency == 0 {
metrics.AverageLatency = latency
} else {
// Exponential moving average
metrics.AverageLatency = time.Duration(float64(metrics.AverageLatency)*0.9 + float64(latency)*0.1)
}
if latency > metrics.MaxLatency {
metrics.MaxLatency = latency
}
// Track latency spikes
if latency > dhm.latencyThreshold {
atomic.AddInt64(&metrics.LatencySpikes, 1)
}
}
// RecordUnderrun records an audio underrun event
func (dhm *DeviceHealthMonitor) RecordUnderrun(deviceType string) {
if atomic.LoadInt32(&dhm.monitoringEnabled) == 0 {
return
}
dhm.mutex.Lock()
defer dhm.mutex.Unlock()
var metrics *DeviceHealthMetrics
switch deviceType {
case "capture":
metrics = dhm.captureMetrics
case "playback":
metrics = dhm.playbackMetrics
default:
return
}
atomic.AddInt64(&metrics.Underruns, 1)
dhm.logger.Debug().Str("device_type", deviceType).Msg("recorded audio underrun")
}
// RecordOverrun records an audio overrun event
func (dhm *DeviceHealthMonitor) RecordOverrun(deviceType string) {
if atomic.LoadInt32(&dhm.monitoringEnabled) == 0 {
return
}
dhm.mutex.Lock()
defer dhm.mutex.Unlock()
var metrics *DeviceHealthMetrics
switch deviceType {
case "capture":
metrics = dhm.captureMetrics
case "playback":
metrics = dhm.playbackMetrics
default:
return
}
atomic.AddInt64(&metrics.Overruns, 1)
dhm.logger.Debug().Str("device_type", deviceType).Msg("recorded audio overrun")
}
// GetHealthMetrics returns current health metrics
func (dhm *DeviceHealthMonitor) GetHealthMetrics() (capture, playback DeviceHealthMetrics) {
dhm.mutex.RLock()
defer dhm.mutex.RUnlock()
return *dhm.captureMetrics, *dhm.playbackMetrics
}
// monitoringLoop runs the main health monitoring loop
func (dhm *DeviceHealthMonitor) monitoringLoop() {
defer close(dhm.doneChan)
ticker := time.NewTicker(dhm.checkInterval)
defer ticker.Stop()
for {
select {
case <-dhm.stopChan:
return
case <-dhm.ctx.Done():
return
case <-ticker.C:
dhm.performHealthCheck()
}
}
}
// performHealthCheck performs a comprehensive health check
func (dhm *DeviceHealthMonitor) performHealthCheck() {
dhm.mutex.Lock()
defer dhm.mutex.Unlock()
// Assess health for both devices
dhm.assessDeviceHealth("capture", dhm.captureMetrics)
dhm.assessDeviceHealth("playback", dhm.playbackMetrics)
// Check if recovery is needed
dhm.checkRecoveryNeeded("capture", dhm.captureMetrics)
dhm.checkRecoveryNeeded("playback", dhm.playbackMetrics)
}
// assessDeviceHealth assesses the health status of a device
func (dhm *DeviceHealthMonitor) assessDeviceHealth(deviceType string, metrics *DeviceHealthMetrics) {
previousStatus := metrics.CurrentStatus
newStatus := dhm.calculateHealthStatus(metrics)
if newStatus != previousStatus {
metrics.CurrentStatus = newStatus
metrics.StatusLastChanged = time.Now()
dhm.logger.Info().
Str("device_type", deviceType).
Str("previous_status", previousStatus.String()).
Str("new_status", newStatus.String()).
Float64("health_score", metrics.HealthScore).
Msg("device health status changed")
}
// Update health score
metrics.HealthScore = dhm.calculateHealthScore(metrics)
}
// calculateHealthStatus determines health status based on metrics
func (dhm *DeviceHealthMonitor) calculateHealthStatus(metrics *DeviceHealthMetrics) DeviceHealthStatus {
consecutiveErrors := atomic.LoadInt64(&metrics.ConsecutiveErrors)
totalErrors := atomic.LoadInt64(&metrics.TotalErrors)
// Critical: Too many consecutive errors or device disconnected recently
if consecutiveErrors >= int64(dhm.recoveryThreshold) {
return DeviceHealthCritical
}
// Critical: No successful operations in a long time
if !metrics.LastSuccessfulOp.IsZero() && time.Since(metrics.LastSuccessfulOp) > time.Duration(dhm.config.SupervisorTimeout) {
return DeviceHealthCritical
}
// Failing: High error rate or frequent latency spikes
if metrics.ErrorRate > dhm.errorRateLimit || atomic.LoadInt64(&metrics.LatencySpikes) > int64(dhm.config.MaxDroppedFrames) {
return DeviceHealthFailing
}
// Degraded: Some errors or performance issues
if consecutiveErrors > 0 || totalErrors > int64(dhm.config.MaxDroppedFrames/2) || metrics.AverageLatency > dhm.latencyThreshold {
return DeviceHealthDegraded
}
// Healthy: No significant issues
return DeviceHealthHealthy
}
// calculateHealthScore calculates a numeric health score (0.0 to 1.0)
func (dhm *DeviceHealthMonitor) calculateHealthScore(metrics *DeviceHealthMetrics) float64 {
score := 1.0
// Penalize consecutive errors
consecutiveErrors := atomic.LoadInt64(&metrics.ConsecutiveErrors)
if consecutiveErrors > 0 {
score -= float64(consecutiveErrors) * 0.1
}
// Penalize high error rate
if metrics.ErrorRate > 0 {
score -= min(0.5, metrics.ErrorRate/dhm.errorRateLimit*0.5)
}
// Penalize high latency
if metrics.AverageLatency > dhm.latencyThreshold {
excess := float64(metrics.AverageLatency-dhm.latencyThreshold) / float64(dhm.latencyThreshold)
score -= min(0.3, excess*0.3)
}
// Penalize underruns/overruns
underruns := atomic.LoadInt64(&metrics.Underruns)
overruns := atomic.LoadInt64(&metrics.Overruns)
if underruns+overruns > 0 {
score -= min(0.2, float64(underruns+overruns)*0.01)
}
return max(0.0, score)
}
// checkRecoveryNeeded checks if recovery is needed and triggers it
func (dhm *DeviceHealthMonitor) checkRecoveryNeeded(deviceType string, metrics *DeviceHealthMetrics) {
if metrics.CurrentStatus == DeviceHealthCritical {
dhm.triggerRecovery(deviceType, metrics)
}
}
// triggerRecovery triggers recovery for a device
func (dhm *DeviceHealthMonitor) triggerRecovery(deviceType string, metrics *DeviceHealthMetrics) {
atomic.AddInt64(&metrics.RecoveryAttempts, 1)
dhm.logger.Warn().
Str("device_type", deviceType).
Str("status", metrics.CurrentStatus.String()).
Int64("consecutive_errors", atomic.LoadInt64(&metrics.ConsecutiveErrors)).
Float64("error_rate", metrics.ErrorRate).
Msg("triggering device recovery")
// Try registered recovery callbacks
dhm.callbackMutex.RLock()
defer dhm.callbackMutex.RUnlock()
for component, callback := range dhm.recoveryCallbacks {
if callback != nil {
go func(comp string, cb func() error) {
if err := cb(); err != nil {
dhm.logger.Error().
Str("component", comp).
Str("device_type", deviceType).
Err(err).
Msg("recovery callback failed")
} else {
atomic.AddInt64(&metrics.SuccessfulRecoveries, 1)
dhm.logger.Info().
Str("component", comp).
Str("device_type", deviceType).
Msg("recovery callback succeeded")
}
}(component, callback)
}
}
}
// Global device health monitor instance
var (
globalDeviceHealthMonitor *DeviceHealthMonitor
deviceHealthOnce sync.Once
)
// GetDeviceHealthMonitor returns the global device health monitor
func GetDeviceHealthMonitor() *DeviceHealthMonitor {
deviceHealthOnce.Do(func() {
globalDeviceHealthMonitor = NewDeviceHealthMonitor()
})
return globalDeviceHealthMonitor
}
// Helper functions for min/max
func min(a, b float64) float64 {
if a < b {
return a
}
return b
}
func max(a, b float64) float64 {
if a > b {
return a
}
return b
}

View File

@ -111,7 +111,7 @@ func initializeBroadcaster() {
go audioEventBroadcaster.startMetricsBroadcasting()
// Start granular metrics logging with same interval as metrics broadcasting
StartGranularMetricsLogging(GetMetricsUpdateInterval())
// StartGranularMetricsLogging(GetMetricsUpdateInterval()) // Disabled to reduce log pollution
}
// InitializeAudioEventBroadcaster initializes the global audio event broadcaster

View File

@ -93,18 +93,18 @@ type BufferPoolEfficiencyTracker struct {
// NewLatencyHistogram creates a new latency histogram with predefined buckets
func NewLatencyHistogram(maxSamples int, logger zerolog.Logger) *LatencyHistogram {
// Define latency buckets: 1ms, 5ms, 10ms, 25ms, 50ms, 100ms, 250ms, 500ms, 1s, 2s+
// Define latency buckets using configuration constants
buckets := []int64{
int64(1 * time.Millisecond),
int64(5 * time.Millisecond),
int64(10 * time.Millisecond),
int64(25 * time.Millisecond),
int64(50 * time.Millisecond),
int64(100 * time.Millisecond),
int64(250 * time.Millisecond),
int64(500 * time.Millisecond),
int64(1 * time.Second),
int64(2 * time.Second),
int64(GetConfig().LatencyBucket10ms),
int64(GetConfig().LatencyBucket25ms),
int64(GetConfig().LatencyBucket50ms),
int64(GetConfig().LatencyBucket100ms),
int64(GetConfig().LatencyBucket250ms),
int64(GetConfig().LatencyBucket500ms),
int64(GetConfig().LatencyBucket1s),
int64(GetConfig().LatencyBucket2s),
}
return &LatencyHistogram{

View File

@ -1,6 +1,7 @@
package audio
import (
"fmt"
"sync/atomic"
"time"
@ -10,10 +11,10 @@ import (
// AudioInputMetrics holds metrics for microphone input
type AudioInputMetrics struct {
FramesSent int64
FramesDropped int64
BytesProcessed int64
ConnectionDrops int64
FramesSent int64 // Total frames sent
FramesDropped int64 // Total frames dropped
BytesProcessed int64 // Total bytes processed
ConnectionDrops int64 // Connection drops
AverageLatency time.Duration // time.Duration is int64
LastFrameTime time.Time
}
@ -31,26 +32,30 @@ type AudioInputManager struct {
func NewAudioInputManager() *AudioInputManager {
return &AudioInputManager{
ipcManager: NewAudioInputIPCManager(),
logger: logging.GetDefaultLogger().With().Str("component", "audio-input").Logger(),
logger: logging.GetDefaultLogger().With().Str("component", AudioInputManagerComponent).Logger(),
}
}
// Start begins processing microphone input
func (aim *AudioInputManager) Start() error {
if !atomic.CompareAndSwapInt32(&aim.running, 0, 1) {
return nil // Already running
return fmt.Errorf("audio input manager is already running")
}
aim.logger.Info().Msg("Starting audio input manager")
aim.logger.Info().Str("component", AudioInputManagerComponent).Msg("starting component")
// Start the IPC-based audio input
err := aim.ipcManager.Start()
if err != nil {
aim.logger.Error().Err(err).Msg("Failed to start IPC audio input")
aim.logger.Error().Err(err).Str("component", AudioInputManagerComponent).Msg("failed to start component")
// Ensure proper cleanup on error
atomic.StoreInt32(&aim.running, 0)
// Reset metrics on failed start
aim.resetMetrics()
return err
}
aim.logger.Info().Str("component", AudioInputManagerComponent).Msg("component started successfully")
return nil
}
@ -60,12 +65,20 @@ func (aim *AudioInputManager) Stop() {
return // Already stopped
}
aim.logger.Info().Msg("Stopping audio input manager")
aim.logger.Info().Str("component", AudioInputManagerComponent).Msg("stopping component")
// Stop the IPC-based audio input
aim.ipcManager.Stop()
aim.logger.Info().Msg("Audio input manager stopped")
aim.logger.Info().Str("component", AudioInputManagerComponent).Msg("component stopped")
}
// resetMetrics resets all metrics to zero
func (aim *AudioInputManager) resetMetrics() {
atomic.StoreInt64(&aim.metrics.FramesSent, 0)
atomic.StoreInt64(&aim.metrics.FramesDropped, 0)
atomic.StoreInt64(&aim.metrics.BytesProcessed, 0)
atomic.StoreInt64(&aim.metrics.ConnectionDrops, 0)
}
// WriteOpusFrame writes an Opus frame to the audio input system with latency tracking

View File

@ -1,7 +1,6 @@
package audio
import (
"context"
"encoding/binary"
"fmt"
"io"
@ -14,12 +13,12 @@ import (
"time"
"github.com/jetkvm/kvm/internal/logging"
"github.com/rs/zerolog"
)
var (
inputMagicNumber uint32 = GetConfig().InputMagicNumber // "JKMI" (JetKVM Microphone Input)
inputSocketName = "audio_input.sock"
writeTimeout = GetConfig().WriteTimeout // Non-blocking write timeout
)
const (
@ -51,6 +50,27 @@ type InputIPCMessage struct {
Data []byte
}
// Implement IPCMessage interface
func (msg *InputIPCMessage) GetMagic() uint32 {
return msg.Magic
}
func (msg *InputIPCMessage) GetType() uint8 {
return uint8(msg.Type)
}
func (msg *InputIPCMessage) GetLength() uint32 {
return msg.Length
}
func (msg *InputIPCMessage) GetTimestamp() int64 {
return msg.Timestamp
}
func (msg *InputIPCMessage) GetData() []byte {
return msg.Data
}
// OptimizedIPCMessage represents an optimized message with pre-allocated buffers
type OptimizedIPCMessage struct {
header [headerSize]byte // Pre-allocated header buffer
@ -80,16 +100,15 @@ var globalMessagePool = &MessagePool{
var messagePoolInitOnce sync.Once
// initializeMessagePool initializes the message pool with pre-allocated messages
// initializeMessagePool initializes the global message pool with pre-allocated messages
func initializeMessagePool() {
messagePoolInitOnce.Do(func() {
// Pre-allocate 30% of pool size for immediate availability
preallocSize := messagePoolSize * GetConfig().InputPreallocPercentage / 100
preallocSize := messagePoolSize / 4 // 25% pre-allocated for immediate use
globalMessagePool.preallocSize = preallocSize
globalMessagePool.maxPoolSize = messagePoolSize * GetConfig().PoolGrowthMultiplier // Allow growth up to 2x
globalMessagePool.preallocated = make([]*OptimizedIPCMessage, 0, preallocSize)
// Pre-allocate messages to reduce initial allocation overhead
// Pre-allocate messages for immediate use
for i := 0; i < preallocSize; i++ {
msg := &OptimizedIPCMessage{
data: make([]byte, 0, maxFrameSize),
@ -97,7 +116,7 @@ func initializeMessagePool() {
globalMessagePool.preallocated = append(globalMessagePool.preallocated, msg)
}
// Fill the channel pool with remaining messages
// Fill the channel with remaining messages
for i := preallocSize; i < messagePoolSize; i++ {
globalMessagePool.pool <- &OptimizedIPCMessage{
data: make([]byte, 0, maxFrameSize),
@ -167,7 +186,7 @@ type InputIPCConfig struct {
// AudioInputServer handles IPC communication for audio input processing
type AudioInputServer struct {
// Atomic fields must be first for proper alignment on ARM
// Atomic fields MUST be first for ARM32 alignment (int64 fields need 8-byte alignment)
bufferSize int64 // Current buffer size (atomic)
processingTime int64 // Average processing time in nanoseconds (atomic)
droppedFrames int64 // Dropped frames counter (atomic)
@ -227,6 +246,11 @@ func (ais *AudioInputServer) Start() error {
ais.running = true
// Reset counters on start
atomic.StoreInt64(&ais.totalFrames, 0)
atomic.StoreInt64(&ais.droppedFrames, 0)
atomic.StoreInt64(&ais.processingTime, 0)
// Start triple-goroutine architecture
ais.startReaderGoroutine()
ais.startProcessorGoroutine()
@ -276,7 +300,9 @@ func (ais *AudioInputServer) acceptConnections() {
conn, err := ais.listener.Accept()
if err != nil {
if ais.running {
// Only log error if we're still supposed to be running
// Log error and continue accepting
logger := logging.GetDefaultLogger().With().Str("component", "audio-input-server").Logger()
logger.Warn().Err(err).Msg("Failed to accept connection, retrying")
continue
}
return
@ -293,9 +319,10 @@ func (ais *AudioInputServer) acceptConnections() {
}
ais.mtx.Lock()
// Close existing connection if any
// Close existing connection if any to prevent resource leaks
if ais.conn != nil {
ais.conn.Close()
ais.conn = nil
}
ais.conn = conn
ais.mtx.Unlock()
@ -461,33 +488,13 @@ func (ais *AudioInputServer) sendAck() error {
return ais.writeMessage(ais.conn, msg)
}
// writeMessage writes a message to the connection using optimized buffers
// Global shared message pool for input IPC server
var globalInputServerMessagePool = NewGenericMessagePool(messagePoolSize)
// writeMessage writes a message to the connection using shared common utilities
func (ais *AudioInputServer) writeMessage(conn net.Conn, msg *InputIPCMessage) error {
// Get optimized message from pool for header preparation
optMsg := globalMessagePool.Get()
defer globalMessagePool.Put(optMsg)
// Prepare header in pre-allocated buffer
binary.LittleEndian.PutUint32(optMsg.header[0:4], msg.Magic)
optMsg.header[4] = byte(msg.Type)
binary.LittleEndian.PutUint32(optMsg.header[5:9], msg.Length)
binary.LittleEndian.PutUint64(optMsg.header[9:17], uint64(msg.Timestamp))
// Write header
_, err := conn.Write(optMsg.header[:])
if err != nil {
return err
}
// Write data if present
if msg.Length > 0 && msg.Data != nil {
_, err = conn.Write(msg.Data)
if err != nil {
return err
}
}
return nil
// Use shared WriteIPCMessage function with global message pool
return WriteIPCMessage(conn, msg, globalInputServerMessagePool, &ais.droppedFrames)
}
// AudioInputClient handles IPC communication from the main process
@ -515,6 +522,12 @@ func (aic *AudioInputClient) Connect() error {
return nil // Already connected
}
// Ensure clean state before connecting
if aic.conn != nil {
aic.conn.Close()
aic.conn = nil
}
socketPath := getInputSocketPath()
// Try connecting multiple times as the server might not be ready
// Reduced retry count and delay for faster startup
@ -523,6 +536,9 @@ func (aic *AudioInputClient) Connect() error {
if err == nil {
aic.conn = conn
aic.running = true
// Reset frame counters on successful connection
atomic.StoreInt64(&aic.totalFrames, 0)
atomic.StoreInt64(&aic.droppedFrames, 0)
return nil
}
// Exponential backoff starting from config
@ -535,7 +551,10 @@ func (aic *AudioInputClient) Connect() error {
time.Sleep(delay)
}
return fmt.Errorf("failed to connect to audio input server")
// Ensure clean state on connection failure
aic.conn = nil
aic.running = false
return fmt.Errorf("failed to connect to audio input server after 10 attempts")
}
// Disconnect disconnects from the audio input server
@ -667,58 +686,15 @@ func (aic *AudioInputClient) SendHeartbeat() error {
}
// writeMessage writes a message to the server
// Global shared message pool for input IPC clients
var globalInputMessagePool = NewGenericMessagePool(messagePoolSize)
func (aic *AudioInputClient) writeMessage(msg *InputIPCMessage) error {
// Increment total frames counter
atomic.AddInt64(&aic.totalFrames, 1)
// Get optimized message from pool for header preparation
optMsg := globalMessagePool.Get()
defer globalMessagePool.Put(optMsg)
// Prepare header in pre-allocated buffer
binary.LittleEndian.PutUint32(optMsg.header[0:4], msg.Magic)
optMsg.header[4] = byte(msg.Type)
binary.LittleEndian.PutUint32(optMsg.header[5:9], msg.Length)
binary.LittleEndian.PutUint64(optMsg.header[9:17], uint64(msg.Timestamp))
// Use non-blocking write with timeout
ctx, cancel := context.WithTimeout(context.Background(), writeTimeout)
defer cancel()
// Create a channel to signal write completion
done := make(chan error, 1)
go func() {
// Write header using pre-allocated buffer
_, err := aic.conn.Write(optMsg.header[:])
if err != nil {
done <- err
return
}
// Write data if present
if msg.Length > 0 && msg.Data != nil {
_, err = aic.conn.Write(msg.Data)
if err != nil {
done <- err
return
}
}
done <- nil
}()
// Wait for completion or timeout
select {
case err := <-done:
if err != nil {
atomic.AddInt64(&aic.droppedFrames, 1)
return err
}
return nil
case <-ctx.Done():
// Timeout occurred - drop frame to prevent blocking
atomic.AddInt64(&aic.droppedFrames, 1)
return fmt.Errorf("write timeout - frame dropped")
}
// Use shared WriteIPCMessage function with global message pool
return WriteIPCMessage(aic.conn, msg, globalInputMessagePool, &aic.droppedFrames)
}
// IsConnected returns whether the client is connected
@ -730,23 +706,19 @@ func (aic *AudioInputClient) IsConnected() bool {
// GetFrameStats returns frame statistics
func (aic *AudioInputClient) GetFrameStats() (total, dropped int64) {
return atomic.LoadInt64(&aic.totalFrames), atomic.LoadInt64(&aic.droppedFrames)
stats := GetFrameStats(&aic.totalFrames, &aic.droppedFrames)
return stats.Total, stats.Dropped
}
// GetDropRate returns the current frame drop rate as a percentage
func (aic *AudioInputClient) GetDropRate() float64 {
total := atomic.LoadInt64(&aic.totalFrames)
dropped := atomic.LoadInt64(&aic.droppedFrames)
if total == 0 {
return 0.0
}
return float64(dropped) / float64(total) * GetConfig().PercentageMultiplier
stats := GetFrameStats(&aic.totalFrames, &aic.droppedFrames)
return CalculateDropRate(stats)
}
// ResetStats resets frame statistics
func (aic *AudioInputClient) ResetStats() {
atomic.StoreInt64(&aic.totalFrames, 0)
atomic.StoreInt64(&aic.droppedFrames, 0)
ResetFrameStats(&aic.totalFrames, &aic.droppedFrames)
}
// startReaderGoroutine starts the message reader goroutine
@ -754,6 +726,17 @@ func (ais *AudioInputServer) startReaderGoroutine() {
ais.wg.Add(1)
go func() {
defer ais.wg.Done()
// Enhanced error tracking and recovery
var consecutiveErrors int
var lastErrorTime time.Time
maxConsecutiveErrors := GetConfig().MaxConsecutiveErrors
errorResetWindow := GetConfig().RestartWindow // Use existing restart window
baseBackoffDelay := GetConfig().RetryDelay
maxBackoffDelay := GetConfig().MaxRetryDelay
logger := logging.GetDefaultLogger().With().Str("component", "audio-input-reader").Logger()
for {
select {
case <-ais.stopChan:
@ -762,8 +745,55 @@ func (ais *AudioInputServer) startReaderGoroutine() {
if ais.conn != nil {
msg, err := ais.readMessage(ais.conn)
if err != nil {
continue // Connection error, retry
// Enhanced error handling with progressive backoff
now := time.Now()
// Reset error counter if enough time has passed
if now.Sub(lastErrorTime) > errorResetWindow {
consecutiveErrors = 0
}
consecutiveErrors++
lastErrorTime = now
// Log error with context
logger.Warn().Err(err).
Int("consecutive_errors", consecutiveErrors).
Msg("Failed to read message from input connection")
// Progressive backoff based on error count
if consecutiveErrors > 1 {
backoffDelay := time.Duration(consecutiveErrors-1) * baseBackoffDelay
if backoffDelay > maxBackoffDelay {
backoffDelay = maxBackoffDelay
}
time.Sleep(backoffDelay)
}
// If too many consecutive errors, close connection to force reconnect
if consecutiveErrors >= maxConsecutiveErrors {
logger.Error().
Int("consecutive_errors", consecutiveErrors).
Msg("Too many consecutive read errors, closing connection")
ais.mtx.Lock()
if ais.conn != nil {
ais.conn.Close()
ais.conn = nil
}
ais.mtx.Unlock()
consecutiveErrors = 0 // Reset for next connection
}
continue
}
// Reset error counter on successful read
if consecutiveErrors > 0 {
consecutiveErrors = 0
logger.Info().Msg("Input connection recovered")
}
// Send to message channel with non-blocking write
select {
case ais.messageChan <- msg:
@ -771,7 +801,11 @@ func (ais *AudioInputServer) startReaderGoroutine() {
default:
// Channel full, drop message
atomic.AddInt64(&ais.droppedFrames, 1)
logger.Warn().Msg("Message channel full, dropping frame")
}
} else {
// No connection, wait briefly before checking again
time.Sleep(GetConfig().DefaultSleepDuration)
}
}
}
@ -796,40 +830,105 @@ func (ais *AudioInputServer) startProcessorGoroutine() {
}
}()
// Enhanced error tracking for processing
var processingErrors int
var lastProcessingError time.Time
maxProcessingErrors := GetConfig().MaxConsecutiveErrors
errorResetWindow := GetConfig().RestartWindow
defer ais.wg.Done()
for {
select {
case <-ais.stopChan:
return
case msg := <-ais.messageChan:
// Intelligent frame dropping: prioritize recent frames
if msg.Type == InputMessageTypeOpusFrame {
// Check if processing queue is getting full
queueLen := len(ais.processChan)
bufferSize := int(atomic.LoadInt64(&ais.bufferSize))
// Process message with error handling
start := time.Now()
err := ais.processMessageWithRecovery(msg, logger)
processingTime := time.Since(start)
if queueLen > bufferSize*3/4 {
// Drop oldest frames, keep newest
select {
case <-ais.processChan: // Remove oldest
atomic.AddInt64(&ais.droppedFrames, 1)
default:
}
if err != nil {
// Track processing errors
now := time.Now()
if now.Sub(lastProcessingError) > errorResetWindow {
processingErrors = 0
}
processingErrors++
lastProcessingError = now
logger.Warn().Err(err).
Int("processing_errors", processingErrors).
Dur("processing_time", processingTime).
Msg("Failed to process input message")
// If too many processing errors, drop frames more aggressively
if processingErrors >= maxProcessingErrors {
logger.Error().
Int("processing_errors", processingErrors).
Msg("Too many processing errors, entering aggressive drop mode")
// Clear processing queue to recover
for len(ais.processChan) > 0 {
select {
case <-ais.processChan:
atomic.AddInt64(&ais.droppedFrames, 1)
default:
break
}
}
processingErrors = 0 // Reset after clearing queue
}
continue
}
// Send to processing queue
select {
case ais.processChan <- msg:
default:
// Processing queue full, drop frame
atomic.AddInt64(&ais.droppedFrames, 1)
// Reset error counter on successful processing
if processingErrors > 0 {
processingErrors = 0
logger.Info().Msg("Input processing recovered")
}
// Update processing time metrics
atomic.StoreInt64(&ais.processingTime, processingTime.Nanoseconds())
}
}
}()
}
// processMessageWithRecovery processes a message with enhanced error recovery
func (ais *AudioInputServer) processMessageWithRecovery(msg *InputIPCMessage, logger zerolog.Logger) error {
// Intelligent frame dropping: prioritize recent frames
if msg.Type == InputMessageTypeOpusFrame {
// Check if processing queue is getting full
queueLen := len(ais.processChan)
bufferSize := int(atomic.LoadInt64(&ais.bufferSize))
if queueLen > bufferSize*3/4 {
// Drop oldest frames, keep newest
select {
case <-ais.processChan: // Remove oldest
atomic.AddInt64(&ais.droppedFrames, 1)
logger.Debug().Msg("Dropped oldest frame to make room")
default:
}
}
}
// Send to processing queue with timeout
select {
case ais.processChan <- msg:
return nil
case <-time.After(GetConfig().WriteTimeout):
// Processing queue full and timeout reached, drop frame
atomic.AddInt64(&ais.droppedFrames, 1)
return fmt.Errorf("processing queue timeout")
default:
// Processing queue full, drop frame immediately
atomic.AddInt64(&ais.droppedFrames, 1)
return fmt.Errorf("processing queue full")
}
}
// startMonitorGoroutine starts the performance monitoring goroutine
func (ais *AudioInputServer) startMonitorGoroutine() {
ais.wg.Add(1)

View File

@ -21,7 +21,7 @@ type AudioInputIPCManager struct {
func NewAudioInputIPCManager() *AudioInputIPCManager {
return &AudioInputIPCManager{
supervisor: NewAudioInputSupervisor(),
logger: logging.GetDefaultLogger().With().Str("component", "audio-input-ipc").Logger(),
logger: logging.GetDefaultLogger().With().Str("component", AudioInputIPCComponent).Logger(),
}
}
@ -31,12 +31,15 @@ func (aim *AudioInputIPCManager) Start() error {
return nil
}
aim.logger.Info().Msg("Starting IPC-based audio input system")
aim.logger.Info().Str("component", AudioInputIPCComponent).Msg("starting component")
err := aim.supervisor.Start()
if err != nil {
// Ensure proper cleanup on supervisor start failure
atomic.StoreInt32(&aim.running, 0)
aim.logger.Error().Err(err).Msg("Failed to start audio input supervisor")
// Reset metrics on failed start
aim.resetMetrics()
aim.logger.Error().Err(err).Str("component", AudioInputIPCComponent).Msg("failed to start audio input supervisor")
return err
}
@ -51,10 +54,11 @@ func (aim *AudioInputIPCManager) Start() error {
err = aim.supervisor.SendConfig(config)
if err != nil {
aim.logger.Warn().Err(err).Msg("Failed to send initial config, will retry later")
// Config send failure is not critical, log warning and continue
aim.logger.Warn().Err(err).Str("component", AudioInputIPCComponent).Msg("failed to send initial config, will retry later")
}
aim.logger.Info().Msg("IPC-based audio input system started")
aim.logger.Info().Str("component", AudioInputIPCComponent).Msg("component started successfully")
return nil
}
@ -64,9 +68,17 @@ func (aim *AudioInputIPCManager) Stop() {
return
}
aim.logger.Info().Msg("Stopping IPC-based audio input system")
aim.logger.Info().Str("component", AudioInputIPCComponent).Msg("stopping component")
aim.supervisor.Stop()
aim.logger.Info().Msg("IPC-based audio input system stopped")
aim.logger.Info().Str("component", AudioInputIPCComponent).Msg("component stopped")
}
// resetMetrics resets all metrics to zero
func (aim *AudioInputIPCManager) resetMetrics() {
atomic.StoreInt64(&aim.metrics.FramesSent, 0)
atomic.StoreInt64(&aim.metrics.FramesDropped, 0)
atomic.StoreInt64(&aim.metrics.BytesProcessed, 0)
atomic.StoreInt64(&aim.metrics.ConnectionDrops, 0)
}
// WriteOpusFrame sends an Opus frame to the audio input server via IPC

View File

@ -0,0 +1,277 @@
package audio
import (
"sync"
"testing"
"time"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
// TestAudioInputIPCManager tests the AudioInputIPCManager component
func TestAudioInputIPCManager(t *testing.T) {
tests := []struct {
name string
testFunc func(t *testing.T)
}{
{"Start", testAudioInputIPCManagerStart},
{"Stop", testAudioInputIPCManagerStop},
{"StartStop", testAudioInputIPCManagerStartStop},
{"IsRunning", testAudioInputIPCManagerIsRunning},
{"IsReady", testAudioInputIPCManagerIsReady},
{"GetMetrics", testAudioInputIPCManagerGetMetrics},
{"ConcurrentOperations", testAudioInputIPCManagerConcurrent},
{"MultipleStarts", testAudioInputIPCManagerMultipleStarts},
{"MultipleStops", testAudioInputIPCManagerMultipleStops},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
tt.testFunc(t)
})
}
}
func testAudioInputIPCManagerStart(t *testing.T) {
manager := NewAudioInputIPCManager()
require.NotNil(t, manager)
// Test initial state
assert.False(t, manager.IsRunning())
assert.False(t, manager.IsReady())
// Test start
err := manager.Start()
assert.NoError(t, err)
assert.True(t, manager.IsRunning())
// Cleanup
manager.Stop()
}
func testAudioInputIPCManagerStop(t *testing.T) {
manager := NewAudioInputIPCManager()
require.NotNil(t, manager)
// Start first
err := manager.Start()
require.NoError(t, err)
assert.True(t, manager.IsRunning())
// Test stop
manager.Stop()
assert.False(t, manager.IsRunning())
assert.False(t, manager.IsReady())
}
func testAudioInputIPCManagerStartStop(t *testing.T) {
manager := NewAudioInputIPCManager()
require.NotNil(t, manager)
// Test multiple start/stop cycles
for i := 0; i < 3; i++ {
// Start
err := manager.Start()
assert.NoError(t, err)
assert.True(t, manager.IsRunning())
// Stop
manager.Stop()
assert.False(t, manager.IsRunning())
}
}
func testAudioInputIPCManagerIsRunning(t *testing.T) {
manager := NewAudioInputIPCManager()
require.NotNil(t, manager)
// Initially not running
assert.False(t, manager.IsRunning())
// Start and check
err := manager.Start()
require.NoError(t, err)
assert.True(t, manager.IsRunning())
// Stop and check
manager.Stop()
assert.False(t, manager.IsRunning())
}
func testAudioInputIPCManagerIsReady(t *testing.T) {
manager := NewAudioInputIPCManager()
require.NotNil(t, manager)
// Initially not ready
assert.False(t, manager.IsReady())
// Start and check ready state
err := manager.Start()
require.NoError(t, err)
// Give some time for initialization
time.Sleep(100 * time.Millisecond)
// Stop
manager.Stop()
assert.False(t, manager.IsReady())
}
func testAudioInputIPCManagerGetMetrics(t *testing.T) {
manager := NewAudioInputIPCManager()
require.NotNil(t, manager)
// Test metrics when not running
metrics := manager.GetMetrics()
assert.NotNil(t, metrics)
// Start and test metrics
err := manager.Start()
require.NoError(t, err)
metrics = manager.GetMetrics()
assert.NotNil(t, metrics)
// Cleanup
manager.Stop()
}
func testAudioInputIPCManagerConcurrent(t *testing.T) {
manager := NewAudioInputIPCManager()
require.NotNil(t, manager)
var wg sync.WaitGroup
const numGoroutines = 10
// Test concurrent starts
wg.Add(numGoroutines)
for i := 0; i < numGoroutines; i++ {
go func() {
defer wg.Done()
manager.Start()
}()
}
wg.Wait()
// Should be running
assert.True(t, manager.IsRunning())
// Test concurrent stops
wg.Add(numGoroutines)
for i := 0; i < numGoroutines; i++ {
go func() {
defer wg.Done()
manager.Stop()
}()
}
wg.Wait()
// Should be stopped
assert.False(t, manager.IsRunning())
}
func testAudioInputIPCManagerMultipleStarts(t *testing.T) {
manager := NewAudioInputIPCManager()
require.NotNil(t, manager)
// First start should succeed
err := manager.Start()
assert.NoError(t, err)
assert.True(t, manager.IsRunning())
// Subsequent starts should be no-op
err = manager.Start()
assert.NoError(t, err)
assert.True(t, manager.IsRunning())
err = manager.Start()
assert.NoError(t, err)
assert.True(t, manager.IsRunning())
// Cleanup
manager.Stop()
}
func testAudioInputIPCManagerMultipleStops(t *testing.T) {
manager := NewAudioInputIPCManager()
require.NotNil(t, manager)
// Start first
err := manager.Start()
require.NoError(t, err)
assert.True(t, manager.IsRunning())
// First stop should work
manager.Stop()
assert.False(t, manager.IsRunning())
// Subsequent stops should be no-op
manager.Stop()
assert.False(t, manager.IsRunning())
manager.Stop()
assert.False(t, manager.IsRunning())
}
// TestAudioInputIPCMetrics tests the AudioInputMetrics functionality
func TestAudioInputIPCMetrics(t *testing.T) {
metrics := &AudioInputMetrics{}
// Test initial state
assert.Equal(t, int64(0), metrics.FramesSent)
assert.Equal(t, int64(0), metrics.FramesDropped)
assert.Equal(t, int64(0), metrics.BytesProcessed)
assert.Equal(t, int64(0), metrics.ConnectionDrops)
assert.Equal(t, time.Duration(0), metrics.AverageLatency)
assert.True(t, metrics.LastFrameTime.IsZero())
// Test field assignment
metrics.FramesSent = 50
metrics.FramesDropped = 2
metrics.BytesProcessed = 512
metrics.ConnectionDrops = 1
metrics.AverageLatency = 5 * time.Millisecond
metrics.LastFrameTime = time.Now()
// Verify assignments
assert.Equal(t, int64(50), metrics.FramesSent)
assert.Equal(t, int64(2), metrics.FramesDropped)
assert.Equal(t, int64(512), metrics.BytesProcessed)
assert.Equal(t, int64(1), metrics.ConnectionDrops)
assert.Equal(t, 5*time.Millisecond, metrics.AverageLatency)
assert.False(t, metrics.LastFrameTime.IsZero())
}
// BenchmarkAudioInputIPCManager benchmarks the AudioInputIPCManager operations
func BenchmarkAudioInputIPCManager(b *testing.B) {
b.Run("Start", func(b *testing.B) {
for i := 0; i < b.N; i++ {
manager := NewAudioInputIPCManager()
manager.Start()
manager.Stop()
}
})
b.Run("IsRunning", func(b *testing.B) {
manager := NewAudioInputIPCManager()
manager.Start()
defer manager.Stop()
b.ResetTimer()
for i := 0; i < b.N; i++ {
manager.IsRunning()
}
})
b.Run("GetMetrics", func(b *testing.B) {
manager := NewAudioInputIPCManager()
manager.Start()
defer manager.Stop()
b.ResetTimer()
for i := 0; i < b.N; i++ {
manager.GetMetrics()
}
})
}

View File

@ -168,7 +168,16 @@ func (ais *AudioInputSupervisor) GetProcessMetrics() *ProcessMetrics {
defer ais.mtx.Unlock()
if ais.cmd == nil || ais.cmd.Process == nil {
return nil
// Return default metrics when no process is running
return &ProcessMetrics{
PID: 0,
CPUPercent: 0.0,
MemoryRSS: 0,
MemoryVMS: 0,
MemoryPercent: 0.0,
Timestamp: time.Now(),
ProcessName: "audio-input-server",
}
}
pid := ais.cmd.Process.Pid
@ -178,12 +187,21 @@ func (ais *AudioInputSupervisor) GetProcessMetrics() *ProcessMetrics {
return &metric
}
}
return nil
// Return default metrics if process not found in monitoring
return &ProcessMetrics{
PID: pid,
CPUPercent: 0.0,
MemoryRSS: 0,
MemoryVMS: 0,
MemoryPercent: 0.0,
Timestamp: time.Now(),
ProcessName: "audio-input-server",
}
}
// monitorSubprocess monitors the subprocess and handles unexpected exits
func (ais *AudioInputSupervisor) monitorSubprocess() {
if ais.cmd == nil {
if ais.cmd == nil || ais.cmd.Process == nil {
return
}

View File

@ -0,0 +1,241 @@
package audio
import (
"sync"
"testing"
"time"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func TestNewAudioInputManager(t *testing.T) {
manager := NewAudioInputManager()
assert.NotNil(t, manager)
assert.False(t, manager.IsRunning())
assert.False(t, manager.IsReady())
}
func TestAudioInputManagerStart(t *testing.T) {
manager := NewAudioInputManager()
require.NotNil(t, manager)
// Test successful start
err := manager.Start()
assert.NoError(t, err)
assert.True(t, manager.IsRunning())
// Test starting already running manager
err = manager.Start()
assert.Error(t, err)
assert.Contains(t, err.Error(), "already running")
// Cleanup
manager.Stop()
}
func TestAudioInputManagerStop(t *testing.T) {
manager := NewAudioInputManager()
require.NotNil(t, manager)
// Test stopping non-running manager
manager.Stop()
assert.False(t, manager.IsRunning())
// Start and then stop
err := manager.Start()
require.NoError(t, err)
assert.True(t, manager.IsRunning())
manager.Stop()
assert.False(t, manager.IsRunning())
}
func TestAudioInputManagerIsRunning(t *testing.T) {
manager := NewAudioInputManager()
require.NotNil(t, manager)
// Test initial state
assert.False(t, manager.IsRunning())
// Test after start
err := manager.Start()
require.NoError(t, err)
assert.True(t, manager.IsRunning())
// Test after stop
manager.Stop()
assert.False(t, manager.IsRunning())
}
func TestAudioInputManagerIsReady(t *testing.T) {
manager := NewAudioInputManager()
require.NotNil(t, manager)
// Test initial state
assert.False(t, manager.IsReady())
// Start manager
err := manager.Start()
require.NoError(t, err)
// Give some time for initialization
time.Sleep(100 * time.Millisecond)
// Test ready state (may vary based on implementation)
// Just ensure the method doesn't panic
_ = manager.IsReady()
// Cleanup
manager.Stop()
}
func TestAudioInputManagerGetMetrics(t *testing.T) {
manager := NewAudioInputManager()
require.NotNil(t, manager)
// Test metrics when not running
metrics := manager.GetMetrics()
assert.NotNil(t, metrics)
assert.Equal(t, int64(0), metrics.FramesSent)
assert.Equal(t, int64(0), metrics.FramesDropped)
assert.Equal(t, int64(0), metrics.BytesProcessed)
assert.Equal(t, int64(0), metrics.ConnectionDrops)
// Start and test metrics
err := manager.Start()
require.NoError(t, err)
metrics = manager.GetMetrics()
assert.NotNil(t, metrics)
assert.GreaterOrEqual(t, metrics.FramesSent, int64(0))
assert.GreaterOrEqual(t, metrics.FramesDropped, int64(0))
assert.GreaterOrEqual(t, metrics.BytesProcessed, int64(0))
assert.GreaterOrEqual(t, metrics.ConnectionDrops, int64(0))
// Cleanup
manager.Stop()
}
func TestAudioInputManagerConcurrentOperations(t *testing.T) {
manager := NewAudioInputManager()
require.NotNil(t, manager)
var wg sync.WaitGroup
// Test concurrent start/stop operations
for i := 0; i < 10; i++ {
wg.Add(2)
go func() {
defer wg.Done()
_ = manager.Start()
}()
go func() {
defer wg.Done()
manager.Stop()
}()
}
// Test concurrent metric access
for i := 0; i < 5; i++ {
wg.Add(1)
go func() {
defer wg.Done()
_ = manager.GetMetrics()
}()
}
// Test concurrent status checks
for i := 0; i < 5; i++ {
wg.Add(2)
go func() {
defer wg.Done()
_ = manager.IsRunning()
}()
go func() {
defer wg.Done()
_ = manager.IsReady()
}()
}
wg.Wait()
// Cleanup
manager.Stop()
}
func TestAudioInputManagerMultipleStartStop(t *testing.T) {
manager := NewAudioInputManager()
require.NotNil(t, manager)
// Test multiple start/stop cycles
for i := 0; i < 5; i++ {
err := manager.Start()
assert.NoError(t, err)
assert.True(t, manager.IsRunning())
manager.Stop()
assert.False(t, manager.IsRunning())
}
}
func TestAudioInputMetrics(t *testing.T) {
metrics := &AudioInputMetrics{
FramesSent: 100,
FramesDropped: 5,
BytesProcessed: 1024,
ConnectionDrops: 2,
AverageLatency: time.Millisecond * 10,
LastFrameTime: time.Now(),
}
assert.Equal(t, int64(100), metrics.FramesSent)
assert.Equal(t, int64(5), metrics.FramesDropped)
assert.Equal(t, int64(1024), metrics.BytesProcessed)
assert.Equal(t, int64(2), metrics.ConnectionDrops)
assert.Equal(t, time.Millisecond*10, metrics.AverageLatency)
assert.False(t, metrics.LastFrameTime.IsZero())
}
// Benchmark tests
func BenchmarkAudioInputManager(b *testing.B) {
manager := NewAudioInputManager()
b.Run("Start", func(b *testing.B) {
b.ResetTimer()
for i := 0; i < b.N; i++ {
_ = manager.Start()
manager.Stop()
}
})
b.Run("GetMetrics", func(b *testing.B) {
_ = manager.Start()
defer manager.Stop()
b.ResetTimer()
for i := 0; i < b.N; i++ {
_ = manager.GetMetrics()
}
})
b.Run("IsRunning", func(b *testing.B) {
_ = manager.Start()
defer manager.Stop()
b.ResetTimer()
for i := 0; i < b.N; i++ {
_ = manager.IsRunning()
}
})
b.Run("IsReady", func(b *testing.B) {
_ = manager.Start()
defer manager.Stop()
b.ResetTimer()
for i := 0; i < b.N; i++ {
_ = manager.IsReady()
}
})
}

View File

@ -1,7 +1,6 @@
package audio
import (
"context"
"encoding/binary"
"fmt"
"io"
@ -35,7 +34,7 @@ const (
OutputMessageTypeAck
)
// OutputIPCMessage represents an IPC message for audio output
// OutputIPCMessage represents a message sent over IPC
type OutputIPCMessage struct {
Magic uint32
Type OutputMessageType
@ -44,62 +43,32 @@ type OutputIPCMessage struct {
Data []byte
}
// OutputOptimizedMessage represents a pre-allocated message for zero-allocation operations
type OutputOptimizedMessage struct {
header [17]byte // Pre-allocated header buffer (using constant value since array size must be compile-time constant)
data []byte // Reusable data buffer
// Implement IPCMessage interface
func (msg *OutputIPCMessage) GetMagic() uint32 {
return msg.Magic
}
// OutputMessagePool manages pre-allocated messages for zero-allocation IPC
type OutputMessagePool struct {
pool chan *OutputOptimizedMessage
func (msg *OutputIPCMessage) GetType() uint8 {
return uint8(msg.Type)
}
// NewOutputMessagePool creates a new message pool
func NewOutputMessagePool(size int) *OutputMessagePool {
pool := &OutputMessagePool{
pool: make(chan *OutputOptimizedMessage, size),
}
// Pre-allocate messages
for i := 0; i < size; i++ {
msg := &OutputOptimizedMessage{
data: make([]byte, GetConfig().OutputMaxFrameSize),
}
pool.pool <- msg
}
return pool
func (msg *OutputIPCMessage) GetLength() uint32 {
return msg.Length
}
// Get retrieves a message from the pool
func (p *OutputMessagePool) Get() *OutputOptimizedMessage {
select {
case msg := <-p.pool:
return msg
default:
// Pool exhausted, create new message
return &OutputOptimizedMessage{
data: make([]byte, GetConfig().OutputMaxFrameSize),
}
}
func (msg *OutputIPCMessage) GetTimestamp() int64 {
return msg.Timestamp
}
// Put returns a message to the pool
func (p *OutputMessagePool) Put(msg *OutputOptimizedMessage) {
select {
case p.pool <- msg:
// Successfully returned to pool
default:
// Pool full, let GC handle it
}
func (msg *OutputIPCMessage) GetData() []byte {
return msg.Data
}
// Global message pool for output IPC
var globalOutputMessagePool = NewOutputMessagePool(GetConfig().OutputMessagePoolSize)
// Global shared message pool for output IPC client header reading
var globalOutputClientMessagePool = NewGenericMessagePool(GetConfig().OutputMessagePoolSize)
type AudioServer struct {
// Atomic fields must be first for proper alignment on ARM
type AudioOutputServer struct {
// Atomic fields MUST be first for ARM32 alignment (int64 fields need 8-byte alignment)
bufferSize int64 // Current buffer size (atomic)
droppedFrames int64 // Dropped frames counter (atomic)
totalFrames int64 // Total frames counter (atomic)
@ -122,7 +91,7 @@ type AudioServer struct {
socketBufferConfig SocketBufferConfig
}
func NewAudioServer() (*AudioServer, error) {
func NewAudioOutputServer() (*AudioOutputServer, error) {
socketPath := getOutputSocketPath()
// Remove existing socket if any
os.Remove(socketPath)
@ -151,7 +120,7 @@ func NewAudioServer() (*AudioServer, error) {
// Initialize socket buffer configuration
socketBufferConfig := DefaultSocketBufferConfig()
return &AudioServer{
return &AudioOutputServer{
listener: listener,
messageChan: make(chan *OutputIPCMessage, initialBufferSize),
stopChan: make(chan struct{}),
@ -162,7 +131,7 @@ func NewAudioServer() (*AudioServer, error) {
}, nil
}
func (s *AudioServer) Start() error {
func (s *AudioOutputServer) Start() error {
s.mtx.Lock()
defer s.mtx.Unlock()
@ -190,12 +159,14 @@ func (s *AudioServer) Start() error {
}
// acceptConnections accepts incoming connections
func (s *AudioServer) acceptConnections() {
func (s *AudioOutputServer) acceptConnections() {
logger := logging.GetDefaultLogger().With().Str("component", "audio-server").Logger()
for s.running {
conn, err := s.listener.Accept()
if err != nil {
if s.running {
// Only log error if we're still supposed to be running
// Log warning and retry on accept failure
logger.Warn().Err(err).Msg("Failed to accept connection, retrying")
continue
}
return
@ -204,7 +175,6 @@ func (s *AudioServer) acceptConnections() {
// Configure socket buffers for optimal performance
if err := ConfigureSocketBuffers(conn, s.socketBufferConfig); err != nil {
// Log warning but don't fail - socket buffer optimization is not critical
logger := logging.GetDefaultLogger().With().Str("component", "audio-server").Logger()
logger.Warn().Err(err).Msg("Failed to configure socket buffers, continuing with defaults")
} else {
// Record socket buffer metrics for monitoring
@ -215,6 +185,7 @@ func (s *AudioServer) acceptConnections() {
// Close existing connection if any
if s.conn != nil {
s.conn.Close()
s.conn = nil
}
s.conn = conn
s.mtx.Unlock()
@ -222,7 +193,7 @@ func (s *AudioServer) acceptConnections() {
}
// startProcessorGoroutine starts the message processor
func (s *AudioServer) startProcessorGoroutine() {
func (s *AudioOutputServer) startProcessorGoroutine() {
s.wg.Add(1)
go func() {
defer s.wg.Done()
@ -243,7 +214,7 @@ func (s *AudioServer) startProcessorGoroutine() {
}()
}
func (s *AudioServer) Stop() {
func (s *AudioOutputServer) Stop() {
s.mtx.Lock()
defer s.mtx.Unlock()
@ -271,7 +242,7 @@ func (s *AudioServer) Stop() {
}
}
func (s *AudioServer) Close() error {
func (s *AudioOutputServer) Close() error {
s.Stop()
if s.listener != nil {
s.listener.Close()
@ -281,7 +252,7 @@ func (s *AudioServer) Close() error {
return nil
}
func (s *AudioServer) SendFrame(frame []byte) error {
func (s *AudioOutputServer) SendFrame(frame []byte) error {
maxFrameSize := GetConfig().OutputMaxFrameSize
if len(frame) > maxFrameSize {
return fmt.Errorf("output frame size validation failed: got %d bytes, maximum allowed %d bytes", len(frame), maxFrameSize)
@ -318,7 +289,10 @@ func (s *AudioServer) SendFrame(frame []byte) error {
}
// sendFrameToClient sends frame data directly to the connected client
func (s *AudioServer) sendFrameToClient(frame []byte) error {
// Global shared message pool for output IPC server
var globalOutputServerMessagePool = NewGenericMessagePool(GetConfig().OutputMessagePoolSize)
func (s *AudioOutputServer) sendFrameToClient(frame []byte) error {
s.mtx.Lock()
defer s.mtx.Unlock()
@ -328,84 +302,55 @@ func (s *AudioServer) sendFrameToClient(frame []byte) error {
start := time.Now()
// Get optimized message from pool
optMsg := globalOutputMessagePool.Get()
defer globalOutputMessagePool.Put(optMsg)
// Prepare header in pre-allocated buffer
binary.LittleEndian.PutUint32(optMsg.header[0:4], outputMagicNumber)
optMsg.header[4] = byte(OutputMessageTypeOpusFrame)
binary.LittleEndian.PutUint32(optMsg.header[5:9], uint32(len(frame)))
binary.LittleEndian.PutUint64(optMsg.header[9:17], uint64(start.UnixNano()))
// Use non-blocking write with timeout
ctx, cancel := context.WithTimeout(context.Background(), GetConfig().OutputWriteTimeout)
defer cancel()
// Create a channel to signal write completion
done := make(chan error, 1)
go func() {
// Write header using pre-allocated buffer
_, err := s.conn.Write(optMsg.header[:])
if err != nil {
done <- err
return
}
// Write frame data
if len(frame) > 0 {
_, err = s.conn.Write(frame)
if err != nil {
done <- err
return
}
}
done <- nil
}()
// Wait for completion or timeout
select {
case err := <-done:
if err != nil {
atomic.AddInt64(&s.droppedFrames, 1)
return err
}
// Record latency for monitoring
if s.latencyMonitor != nil {
writeLatency := time.Since(start)
s.latencyMonitor.RecordLatency(writeLatency, "ipc_write")
}
return nil
case <-ctx.Done():
// Timeout occurred - drop frame to prevent blocking
atomic.AddInt64(&s.droppedFrames, 1)
return fmt.Errorf("write timeout after %v - frame dropped to prevent blocking", GetConfig().OutputWriteTimeout)
// Create output IPC message
msg := &OutputIPCMessage{
Magic: outputMagicNumber,
Type: OutputMessageTypeOpusFrame,
Length: uint32(len(frame)),
Timestamp: start.UnixNano(),
Data: frame,
}
// Use shared WriteIPCMessage function
err := WriteIPCMessage(s.conn, msg, globalOutputServerMessagePool, &s.droppedFrames)
if err != nil {
return err
}
// Record latency for monitoring
if s.latencyMonitor != nil {
writeLatency := time.Since(start)
s.latencyMonitor.RecordLatency(writeLatency, "ipc_write")
}
return nil
}
// GetServerStats returns server performance statistics
func (s *AudioServer) GetServerStats() (total, dropped int64, bufferSize int64) {
return atomic.LoadInt64(&s.totalFrames),
atomic.LoadInt64(&s.droppedFrames),
atomic.LoadInt64(&s.bufferSize)
func (s *AudioOutputServer) GetServerStats() (total, dropped int64, bufferSize int64) {
stats := GetFrameStats(&s.totalFrames, &s.droppedFrames)
return stats.Total, stats.Dropped, atomic.LoadInt64(&s.bufferSize)
}
type AudioClient struct {
// Atomic fields must be first for proper alignment on ARM
type AudioOutputClient struct {
// Atomic fields MUST be first for ARM32 alignment (int64 fields need 8-byte alignment)
droppedFrames int64 // Atomic counter for dropped frames
totalFrames int64 // Atomic counter for total frames
conn net.Conn
mtx sync.Mutex
running bool
conn net.Conn
mtx sync.Mutex
running bool
bufferPool *AudioBufferPool // Buffer pool for memory optimization
}
func NewAudioClient() *AudioClient {
return &AudioClient{}
func NewAudioOutputClient() *AudioOutputClient {
return &AudioOutputClient{
bufferPool: NewAudioBufferPool(GetMaxAudioFrameSize()),
}
}
// Connect connects to the audio output server
func (c *AudioClient) Connect() error {
func (c *AudioOutputClient) Connect() error {
c.mtx.Lock()
defer c.mtx.Unlock()
@ -437,7 +382,7 @@ func (c *AudioClient) Connect() error {
}
// Disconnect disconnects from the audio output server
func (c *AudioClient) Disconnect() {
func (c *AudioOutputClient) Disconnect() {
c.mtx.Lock()
defer c.mtx.Unlock()
@ -453,18 +398,18 @@ func (c *AudioClient) Disconnect() {
}
// IsConnected returns whether the client is connected
func (c *AudioClient) IsConnected() bool {
func (c *AudioOutputClient) IsConnected() bool {
c.mtx.Lock()
defer c.mtx.Unlock()
return c.running && c.conn != nil
}
func (c *AudioClient) Close() error {
func (c *AudioOutputClient) Close() error {
c.Disconnect()
return nil
}
func (c *AudioClient) ReceiveFrame() ([]byte, error) {
func (c *AudioOutputClient) ReceiveFrame() ([]byte, error) {
c.mtx.Lock()
defer c.mtx.Unlock()
@ -473,8 +418,8 @@ func (c *AudioClient) ReceiveFrame() ([]byte, error) {
}
// Get optimized message from pool for header reading
optMsg := globalOutputMessagePool.Get()
defer globalOutputMessagePool.Put(optMsg)
optMsg := globalOutputClientMessagePool.Get()
defer globalOutputClientMessagePool.Put(optMsg)
// Read header
if _, err := io.ReadFull(c.conn, optMsg.header[:]); err != nil {
@ -498,22 +443,26 @@ func (c *AudioClient) ReceiveFrame() ([]byte, error) {
return nil, fmt.Errorf("received frame size validation failed: got %d bytes, maximum allowed %d bytes", size, maxFrameSize)
}
// Read frame data
frame := make([]byte, size)
// Read frame data using buffer pool to avoid allocation
frame := c.bufferPool.Get()
frame = frame[:size] // Resize to actual frame size
if size > 0 {
if _, err := io.ReadFull(c.conn, frame); err != nil {
c.bufferPool.Put(frame) // Return buffer on error
return nil, fmt.Errorf("failed to read frame data: %w", err)
}
}
// Note: Caller is responsible for returning frame to pool via PutAudioFrameBuffer()
atomic.AddInt64(&c.totalFrames, 1)
return frame, nil
}
// GetClientStats returns client performance statistics
func (c *AudioClient) GetClientStats() (total, dropped int64) {
return atomic.LoadInt64(&c.totalFrames),
atomic.LoadInt64(&c.droppedFrames)
func (c *AudioOutputClient) GetClientStats() (total, dropped int64) {
stats := GetFrameStats(&c.totalFrames, &c.droppedFrames)
return stats.Total, stats.Dropped
}
// Helper functions

View File

@ -0,0 +1,238 @@
package audio
import (
"context"
"encoding/binary"
"fmt"
"net"
"sync"
"sync/atomic"
"time"
)
// Common IPC message interface
type IPCMessage interface {
GetMagic() uint32
GetType() uint8
GetLength() uint32
GetTimestamp() int64
GetData() []byte
}
// Common optimized message structure
type OptimizedMessage struct {
header [17]byte // Pre-allocated header buffer
data []byte // Reusable data buffer
}
// Generic message pool for both input and output
type GenericMessagePool struct {
// 64-bit fields must be first for proper alignment on ARM
hitCount int64 // Pool hit counter (atomic)
missCount int64 // Pool miss counter (atomic)
pool chan *OptimizedMessage
preallocated []*OptimizedMessage // Pre-allocated messages
preallocSize int
maxPoolSize int
mutex sync.RWMutex
}
// NewGenericMessagePool creates a new generic message pool
func NewGenericMessagePool(size int) *GenericMessagePool {
pool := &GenericMessagePool{
pool: make(chan *OptimizedMessage, size),
preallocSize: size / 4, // 25% pre-allocated for immediate use
maxPoolSize: size,
}
// Pre-allocate some messages for immediate use
pool.preallocated = make([]*OptimizedMessage, pool.preallocSize)
for i := 0; i < pool.preallocSize; i++ {
pool.preallocated[i] = &OptimizedMessage{
data: make([]byte, 0, GetConfig().MaxFrameSize),
}
}
// Fill the channel pool
for i := 0; i < size-pool.preallocSize; i++ {
select {
case pool.pool <- &OptimizedMessage{
data: make([]byte, 0, GetConfig().MaxFrameSize),
}:
default:
break
}
}
return pool
}
// Get retrieves an optimized message from the pool
func (mp *GenericMessagePool) Get() *OptimizedMessage {
// Try pre-allocated first (fastest path)
mp.mutex.Lock()
if len(mp.preallocated) > 0 {
msg := mp.preallocated[len(mp.preallocated)-1]
mp.preallocated = mp.preallocated[:len(mp.preallocated)-1]
mp.mutex.Unlock()
atomic.AddInt64(&mp.hitCount, 1)
return msg
}
mp.mutex.Unlock()
// Try channel pool
select {
case msg := <-mp.pool:
atomic.AddInt64(&mp.hitCount, 1)
return msg
default:
// Pool empty, create new message
atomic.AddInt64(&mp.missCount, 1)
return &OptimizedMessage{
data: make([]byte, 0, GetConfig().MaxFrameSize),
}
}
}
// Put returns an optimized message to the pool
func (mp *GenericMessagePool) Put(msg *OptimizedMessage) {
if msg == nil {
return
}
// Reset the message for reuse
msg.data = msg.data[:0]
// Try to return to pre-allocated slice first
mp.mutex.Lock()
if len(mp.preallocated) < mp.preallocSize {
mp.preallocated = append(mp.preallocated, msg)
mp.mutex.Unlock()
return
}
mp.mutex.Unlock()
// Try to return to channel pool
select {
case mp.pool <- msg:
// Successfully returned to pool
default:
// Pool full, let GC handle it
}
}
// GetStats returns pool statistics
func (mp *GenericMessagePool) GetStats() (hitCount, missCount int64, hitRate float64) {
hits := atomic.LoadInt64(&mp.hitCount)
misses := atomic.LoadInt64(&mp.missCount)
total := hits + misses
if total > 0 {
hitRate = float64(hits) / float64(total) * 100
}
return hits, misses, hitRate
}
// Common write message function
func WriteIPCMessage(conn net.Conn, msg IPCMessage, pool *GenericMessagePool, droppedFramesCounter *int64) error {
if conn == nil {
return fmt.Errorf("connection is nil")
}
// Get optimized message from pool for header preparation
optMsg := pool.Get()
defer pool.Put(optMsg)
// Prepare header in pre-allocated buffer
binary.LittleEndian.PutUint32(optMsg.header[0:4], msg.GetMagic())
optMsg.header[4] = msg.GetType()
binary.LittleEndian.PutUint32(optMsg.header[5:9], msg.GetLength())
binary.LittleEndian.PutUint64(optMsg.header[9:17], uint64(msg.GetTimestamp()))
// Use non-blocking write with timeout
ctx, cancel := context.WithTimeout(context.Background(), GetConfig().WriteTimeout)
defer cancel()
// Create a channel to signal write completion
done := make(chan error, 1)
go func() {
// Write header using pre-allocated buffer
_, err := conn.Write(optMsg.header[:])
if err != nil {
done <- err
return
}
// Write data if present
if msg.GetLength() > 0 && msg.GetData() != nil {
_, err = conn.Write(msg.GetData())
if err != nil {
done <- err
return
}
}
done <- nil
}()
// Wait for completion or timeout
select {
case err := <-done:
if err != nil {
if droppedFramesCounter != nil {
atomic.AddInt64(droppedFramesCounter, 1)
}
return err
}
return nil
case <-ctx.Done():
// Timeout occurred - drop frame to prevent blocking
if droppedFramesCounter != nil {
atomic.AddInt64(droppedFramesCounter, 1)
}
return fmt.Errorf("write timeout - frame dropped")
}
}
// Common connection acceptance with retry logic
func AcceptConnectionWithRetry(listener net.Listener, maxRetries int, retryDelay time.Duration) (net.Conn, error) {
var lastErr error
for i := 0; i < maxRetries; i++ {
conn, err := listener.Accept()
if err == nil {
return conn, nil
}
lastErr = err
if i < maxRetries-1 {
time.Sleep(retryDelay)
}
}
return nil, fmt.Errorf("failed to accept connection after %d retries: %w", maxRetries, lastErr)
}
// Common frame statistics structure
type FrameStats struct {
Total int64
Dropped int64
}
// GetFrameStats safely retrieves frame statistics
func GetFrameStats(totalCounter, droppedCounter *int64) FrameStats {
return FrameStats{
Total: atomic.LoadInt64(totalCounter),
Dropped: atomic.LoadInt64(droppedCounter),
}
}
// CalculateDropRate calculates the drop rate percentage
func CalculateDropRate(stats FrameStats) float64 {
if stats.Total == 0 {
return 0.0
}
return float64(stats.Dropped) / float64(stats.Total) * 100.0
}
// ResetFrameStats resets frame counters
func ResetFrameStats(totalCounter, droppedCounter *int64) {
atomic.StoreInt64(totalCounter, 0)
atomic.StoreInt64(droppedCounter, 0)
}

View File

@ -301,8 +301,45 @@ var (
micConnectionDropsValue int64
)
// UnifiedAudioMetrics provides a common structure for both input and output audio streams
type UnifiedAudioMetrics struct {
FramesReceived int64 `json:"frames_received"`
FramesDropped int64 `json:"frames_dropped"`
FramesSent int64 `json:"frames_sent,omitempty"`
BytesProcessed int64 `json:"bytes_processed"`
ConnectionDrops int64 `json:"connection_drops"`
LastFrameTime time.Time `json:"last_frame_time"`
AverageLatency time.Duration `json:"average_latency"`
}
// convertAudioMetricsToUnified converts AudioMetrics to UnifiedAudioMetrics
func convertAudioMetricsToUnified(metrics AudioMetrics) UnifiedAudioMetrics {
return UnifiedAudioMetrics{
FramesReceived: metrics.FramesReceived,
FramesDropped: metrics.FramesDropped,
FramesSent: 0, // AudioMetrics doesn't have FramesSent
BytesProcessed: metrics.BytesProcessed,
ConnectionDrops: metrics.ConnectionDrops,
LastFrameTime: metrics.LastFrameTime,
AverageLatency: metrics.AverageLatency,
}
}
// convertAudioInputMetricsToUnified converts AudioInputMetrics to UnifiedAudioMetrics
func convertAudioInputMetricsToUnified(metrics AudioInputMetrics) UnifiedAudioMetrics {
return UnifiedAudioMetrics{
FramesReceived: 0, // AudioInputMetrics doesn't have FramesReceived
FramesDropped: metrics.FramesDropped,
FramesSent: metrics.FramesSent,
BytesProcessed: metrics.BytesProcessed,
ConnectionDrops: metrics.ConnectionDrops,
LastFrameTime: metrics.LastFrameTime,
AverageLatency: metrics.AverageLatency,
}
}
// UpdateAudioMetrics updates Prometheus metrics with current audio data
func UpdateAudioMetrics(metrics AudioMetrics) {
func UpdateAudioMetrics(metrics UnifiedAudioMetrics) {
oldReceived := atomic.SwapInt64(&audioFramesReceivedValue, metrics.FramesReceived)
if metrics.FramesReceived > oldReceived {
audioFramesReceivedTotal.Add(float64(metrics.FramesReceived - oldReceived))
@ -333,7 +370,7 @@ func UpdateAudioMetrics(metrics AudioMetrics) {
}
// UpdateMicrophoneMetrics updates Prometheus metrics with current microphone data
func UpdateMicrophoneMetrics(metrics AudioInputMetrics) {
func UpdateMicrophoneMetrics(metrics UnifiedAudioMetrics) {
oldSent := atomic.SwapInt64(&micFramesSentValue, metrics.FramesSent)
if metrics.FramesSent > oldSent {
microphoneFramesSentTotal.Add(float64(metrics.FramesSent - oldSent))
@ -457,11 +494,11 @@ func StartMetricsUpdater() {
for range ticker.C {
// Update audio output metrics
audioMetrics := GetAudioMetrics()
UpdateAudioMetrics(audioMetrics)
UpdateAudioMetrics(convertAudioMetricsToUnified(audioMetrics))
// Update microphone input metrics
micMetrics := GetAudioInputMetrics()
UpdateMicrophoneMetrics(micMetrics)
UpdateMicrophoneMetrics(convertAudioInputMetricsToUnified(micMetrics))
// Update microphone subprocess process metrics
if inputSupervisor := GetAudioInputIPCSupervisor(); inputSupervisor != nil {

View File

@ -0,0 +1,120 @@
package audio
import "time"
// Naming Standards Documentation
// This file documents the standardized naming conventions for audio components
// to ensure consistency across the entire audio system.
/*
STANDARDIZED NAMING CONVENTIONS:
1. COMPONENT HIERARCHY:
- Manager: High-level component that orchestrates multiple subsystems
- Supervisor: Process lifecycle management (start/stop/restart processes)
- Server: IPC server that handles incoming connections
- Client: IPC client that connects to servers
- Streamer: High-performance streaming component
2. NAMING PATTERNS:
Input Components:
- AudioInputManager (replaces: AudioInputManager)
- AudioInputSupervisor (replaces: AudioInputSupervisor)
- AudioInputServer (replaces: AudioInputServer)
- AudioInputClient (replaces: AudioInputClient)
- AudioInputStreamer (new: for consistency with OutputStreamer)
Output Components:
- AudioOutputManager (new: missing high-level manager)
- AudioOutputSupervisor (replaces: AudioOutputSupervisor)
- AudioOutputServer (replaces: AudioOutputServer)
- AudioOutputClient (replaces: AudioOutputClient)
- AudioOutputStreamer (replaces: OutputStreamer)
3. IPC NAMING:
- AudioInputIPCManager (replaces: AudioInputIPCManager)
- AudioOutputIPCManager (new: for consistency)
4. CONFIGURATION NAMING:
- InputIPCConfig (replaces: InputIPCConfig)
- OutputIPCConfig (new: for consistency)
5. MESSAGE NAMING:
- InputIPCMessage (replaces: InputIPCMessage)
- OutputIPCMessage (replaces: OutputIPCMessage)
- InputMessageType (replaces: InputMessageType)
- OutputMessageType (replaces: OutputMessageType)
ISSUES IDENTIFIED:
1. Missing AudioOutputManager (high-level output management)
2. Inconsistent naming: OutputStreamer vs AudioInputSupervisor
3. Missing AudioOutputIPCManager for symmetry
4. Missing OutputIPCConfig for consistency
5. Component names in logging should be standardized
IMPLEMENTATION PLAN:
1. Create AudioOutputManager to match AudioInputManager
2. Rename OutputStreamer to AudioOutputStreamer
3. Create AudioOutputIPCManager for symmetry
4. Standardize all component logging names
5. Update all references consistently
*/
// Component name constants for consistent logging
const (
// Input component names
AudioInputManagerComponent = "audio-input-manager"
AudioInputSupervisorComponent = "audio-input-supervisor"
AudioInputServerComponent = "audio-input-server"
AudioInputClientComponent = "audio-input-client"
AudioInputIPCComponent = "audio-input-ipc"
// Output component names
AudioOutputManagerComponent = "audio-output-manager"
AudioOutputSupervisorComponent = "audio-output-supervisor"
AudioOutputServerComponent = "audio-output-server"
AudioOutputClientComponent = "audio-output-client"
AudioOutputStreamerComponent = "audio-output-streamer"
AudioOutputIPCComponent = "audio-output-ipc"
// Common component names
AudioRelayComponent = "audio-relay"
AudioEventsComponent = "audio-events"
AudioMetricsComponent = "audio-metrics"
)
// Interface definitions for consistent component behavior
type AudioManagerInterface interface {
Start() error
Stop()
IsRunning() bool
IsReady() bool
GetMetrics() interface{}
}
type AudioSupervisorInterface interface {
Start() error
Stop() error
IsRunning() bool
GetProcessPID() int
GetProcessMetrics() *ProcessMetrics
}
type AudioServerInterface interface {
Start() error
Stop()
Close() error
}
type AudioClientInterface interface {
Connect() error
Disconnect()
IsConnected() bool
Close() error
}
type AudioStreamerInterface interface {
Start() error
Stop()
GetStats() (processed, dropped int64, avgProcessingTime time.Duration)
}

View File

@ -0,0 +1,177 @@
package audio
import (
"sync/atomic"
"time"
"github.com/jetkvm/kvm/internal/logging"
"github.com/rs/zerolog"
)
// AudioOutputManager manages audio output stream using IPC mode
type AudioOutputManager struct {
metrics AudioOutputMetrics
streamer *AudioOutputStreamer
logger zerolog.Logger
running int32
}
// AudioOutputMetrics tracks output-specific metrics
type AudioOutputMetrics struct {
FramesReceived int64
FramesDropped int64
BytesProcessed int64
ConnectionDrops int64
LastFrameTime time.Time
AverageLatency time.Duration
}
// NewAudioOutputManager creates a new audio output manager
func NewAudioOutputManager() *AudioOutputManager {
streamer, err := NewAudioOutputStreamer()
if err != nil {
// Log error but continue with nil streamer - will be handled gracefully
logger := logging.GetDefaultLogger().With().Str("component", AudioOutputManagerComponent).Logger()
logger.Error().Err(err).Msg("Failed to create audio output streamer")
}
return &AudioOutputManager{
streamer: streamer,
logger: logging.GetDefaultLogger().With().Str("component", AudioOutputManagerComponent).Logger(),
}
}
// Start starts the audio output manager
func (aom *AudioOutputManager) Start() error {
if !atomic.CompareAndSwapInt32(&aom.running, 0, 1) {
return nil // Already running
}
aom.logger.Info().Str("component", AudioOutputManagerComponent).Msg("starting component")
if aom.streamer == nil {
// Try to recreate streamer if it was nil
streamer, err := NewAudioOutputStreamer()
if err != nil {
atomic.StoreInt32(&aom.running, 0)
aom.logger.Error().Err(err).Str("component", AudioOutputManagerComponent).Msg("failed to create audio output streamer")
return err
}
aom.streamer = streamer
}
err := aom.streamer.Start()
if err != nil {
atomic.StoreInt32(&aom.running, 0)
// Reset metrics on failed start
aom.resetMetrics()
aom.logger.Error().Err(err).Str("component", AudioOutputManagerComponent).Msg("failed to start component")
return err
}
aom.logger.Info().Str("component", AudioOutputManagerComponent).Msg("component started successfully")
return nil
}
// Stop stops the audio output manager
func (aom *AudioOutputManager) Stop() {
if !atomic.CompareAndSwapInt32(&aom.running, 1, 0) {
return // Already stopped
}
aom.logger.Info().Str("component", AudioOutputManagerComponent).Msg("stopping component")
if aom.streamer != nil {
aom.streamer.Stop()
}
aom.logger.Info().Str("component", AudioOutputManagerComponent).Msg("component stopped")
}
// resetMetrics resets all metrics to zero
func (aom *AudioOutputManager) resetMetrics() {
atomic.StoreInt64(&aom.metrics.FramesReceived, 0)
atomic.StoreInt64(&aom.metrics.FramesDropped, 0)
atomic.StoreInt64(&aom.metrics.BytesProcessed, 0)
atomic.StoreInt64(&aom.metrics.ConnectionDrops, 0)
}
// IsRunning returns whether the audio output manager is running
func (aom *AudioOutputManager) IsRunning() bool {
return atomic.LoadInt32(&aom.running) == 1
}
// IsReady returns whether the audio output manager is ready to receive frames
func (aom *AudioOutputManager) IsReady() bool {
if !aom.IsRunning() || aom.streamer == nil {
return false
}
// For output, we consider it ready if the streamer is running
// This could be enhanced with connection status checks
return true
}
// GetMetrics returns current metrics
func (aom *AudioOutputManager) GetMetrics() AudioOutputMetrics {
return AudioOutputMetrics{
FramesReceived: atomic.LoadInt64(&aom.metrics.FramesReceived),
FramesDropped: atomic.LoadInt64(&aom.metrics.FramesDropped),
BytesProcessed: atomic.LoadInt64(&aom.metrics.BytesProcessed),
ConnectionDrops: atomic.LoadInt64(&aom.metrics.ConnectionDrops),
AverageLatency: aom.metrics.AverageLatency,
LastFrameTime: aom.metrics.LastFrameTime,
}
}
// GetComprehensiveMetrics returns detailed performance metrics
func (aom *AudioOutputManager) GetComprehensiveMetrics() map[string]interface{} {
baseMetrics := aom.GetMetrics()
comprehensiveMetrics := map[string]interface{}{
"manager": map[string]interface{}{
"frames_received": baseMetrics.FramesReceived,
"frames_dropped": baseMetrics.FramesDropped,
"bytes_processed": baseMetrics.BytesProcessed,
"connection_drops": baseMetrics.ConnectionDrops,
"average_latency_ms": float64(baseMetrics.AverageLatency.Nanoseconds()) / 1e6,
"last_frame_time": baseMetrics.LastFrameTime,
"running": aom.IsRunning(),
"ready": aom.IsReady(),
},
}
if aom.streamer != nil {
processed, dropped, avgTime := aom.streamer.GetStats()
comprehensiveMetrics["streamer"] = map[string]interface{}{
"frames_processed": processed,
"frames_dropped": dropped,
"avg_processing_time_ms": float64(avgTime.Nanoseconds()) / 1e6,
}
if detailedStats := aom.streamer.GetDetailedStats(); detailedStats != nil {
comprehensiveMetrics["detailed"] = detailedStats
}
}
return comprehensiveMetrics
}
// LogPerformanceStats logs current performance statistics
func (aom *AudioOutputManager) LogPerformanceStats() {
metrics := aom.GetMetrics()
aom.logger.Info().
Int64("frames_received", metrics.FramesReceived).
Int64("frames_dropped", metrics.FramesDropped).
Int64("bytes_processed", metrics.BytesProcessed).
Int64("connection_drops", metrics.ConnectionDrops).
Float64("average_latency_ms", float64(metrics.AverageLatency.Nanoseconds())/1e6).
Bool("running", aom.IsRunning()).
Bool("ready", aom.IsReady()).
Msg("Audio output manager performance stats")
}
// GetStreamer returns the streamer for advanced operations
func (aom *AudioOutputManager) GetStreamer() *AudioOutputStreamer {
return aom.streamer
}

View File

@ -0,0 +1,277 @@
package audio
import (
"sync"
"testing"
"time"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
// TestAudioOutputManager tests the AudioOutputManager component
func TestAudioOutputManager(t *testing.T) {
tests := []struct {
name string
testFunc func(t *testing.T)
}{
{"Start", testAudioOutputManagerStart},
{"Stop", testAudioOutputManagerStop},
{"StartStop", testAudioOutputManagerStartStop},
{"IsRunning", testAudioOutputManagerIsRunning},
{"IsReady", testAudioOutputManagerIsReady},
{"GetMetrics", testAudioOutputManagerGetMetrics},
{"ConcurrentOperations", testAudioOutputManagerConcurrent},
{"MultipleStarts", testAudioOutputManagerMultipleStarts},
{"MultipleStops", testAudioOutputManagerMultipleStops},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
tt.testFunc(t)
})
}
}
func testAudioOutputManagerStart(t *testing.T) {
manager := NewAudioOutputManager()
require.NotNil(t, manager)
// Test initial state
assert.False(t, manager.IsRunning())
assert.False(t, manager.IsReady())
// Test start
err := manager.Start()
assert.NoError(t, err)
assert.True(t, manager.IsRunning())
// Cleanup
manager.Stop()
}
func testAudioOutputManagerStop(t *testing.T) {
manager := NewAudioOutputManager()
require.NotNil(t, manager)
// Start first
err := manager.Start()
require.NoError(t, err)
assert.True(t, manager.IsRunning())
// Test stop
manager.Stop()
assert.False(t, manager.IsRunning())
assert.False(t, manager.IsReady())
}
func testAudioOutputManagerStartStop(t *testing.T) {
manager := NewAudioOutputManager()
require.NotNil(t, manager)
// Test multiple start/stop cycles
for i := 0; i < 3; i++ {
// Start
err := manager.Start()
assert.NoError(t, err)
assert.True(t, manager.IsRunning())
// Stop
manager.Stop()
assert.False(t, manager.IsRunning())
}
}
func testAudioOutputManagerIsRunning(t *testing.T) {
manager := NewAudioOutputManager()
require.NotNil(t, manager)
// Initially not running
assert.False(t, manager.IsRunning())
// Start and check
err := manager.Start()
require.NoError(t, err)
assert.True(t, manager.IsRunning())
// Stop and check
manager.Stop()
assert.False(t, manager.IsRunning())
}
func testAudioOutputManagerIsReady(t *testing.T) {
manager := NewAudioOutputManager()
require.NotNil(t, manager)
// Initially not ready
assert.False(t, manager.IsReady())
// Start and check ready state
err := manager.Start()
require.NoError(t, err)
// Give some time for initialization
time.Sleep(100 * time.Millisecond)
// Stop
manager.Stop()
assert.False(t, manager.IsReady())
}
func testAudioOutputManagerGetMetrics(t *testing.T) {
manager := NewAudioOutputManager()
require.NotNil(t, manager)
// Test metrics when not running
metrics := manager.GetMetrics()
assert.NotNil(t, metrics)
// Start and test metrics
err := manager.Start()
require.NoError(t, err)
metrics = manager.GetMetrics()
assert.NotNil(t, metrics)
// Cleanup
manager.Stop()
}
func testAudioOutputManagerConcurrent(t *testing.T) {
manager := NewAudioOutputManager()
require.NotNil(t, manager)
var wg sync.WaitGroup
const numGoroutines = 10
// Test concurrent starts
wg.Add(numGoroutines)
for i := 0; i < numGoroutines; i++ {
go func() {
defer wg.Done()
manager.Start()
}()
}
wg.Wait()
// Should be running
assert.True(t, manager.IsRunning())
// Test concurrent stops
wg.Add(numGoroutines)
for i := 0; i < numGoroutines; i++ {
go func() {
defer wg.Done()
manager.Stop()
}()
}
wg.Wait()
// Should be stopped
assert.False(t, manager.IsRunning())
}
func testAudioOutputManagerMultipleStarts(t *testing.T) {
manager := NewAudioOutputManager()
require.NotNil(t, manager)
// First start should succeed
err := manager.Start()
assert.NoError(t, err)
assert.True(t, manager.IsRunning())
// Subsequent starts should be no-op
err = manager.Start()
assert.NoError(t, err)
assert.True(t, manager.IsRunning())
err = manager.Start()
assert.NoError(t, err)
assert.True(t, manager.IsRunning())
// Cleanup
manager.Stop()
}
func testAudioOutputManagerMultipleStops(t *testing.T) {
manager := NewAudioOutputManager()
require.NotNil(t, manager)
// Start first
err := manager.Start()
require.NoError(t, err)
assert.True(t, manager.IsRunning())
// First stop should work
manager.Stop()
assert.False(t, manager.IsRunning())
// Subsequent stops should be no-op
manager.Stop()
assert.False(t, manager.IsRunning())
manager.Stop()
assert.False(t, manager.IsRunning())
}
// TestAudioOutputMetrics tests the AudioOutputMetrics functionality
func TestAudioOutputMetrics(t *testing.T) {
metrics := &AudioOutputMetrics{}
// Test initial state
assert.Equal(t, int64(0), metrics.FramesReceived)
assert.Equal(t, int64(0), metrics.FramesDropped)
assert.Equal(t, int64(0), metrics.BytesProcessed)
assert.Equal(t, int64(0), metrics.ConnectionDrops)
assert.Equal(t, time.Duration(0), metrics.AverageLatency)
assert.True(t, metrics.LastFrameTime.IsZero())
// Test field assignment
metrics.FramesReceived = 100
metrics.FramesDropped = 5
metrics.BytesProcessed = 1024
metrics.ConnectionDrops = 2
metrics.AverageLatency = 10 * time.Millisecond
metrics.LastFrameTime = time.Now()
// Verify assignments
assert.Equal(t, int64(100), metrics.FramesReceived)
assert.Equal(t, int64(5), metrics.FramesDropped)
assert.Equal(t, int64(1024), metrics.BytesProcessed)
assert.Equal(t, int64(2), metrics.ConnectionDrops)
assert.Equal(t, 10*time.Millisecond, metrics.AverageLatency)
assert.False(t, metrics.LastFrameTime.IsZero())
}
// BenchmarkAudioOutputManager benchmarks the AudioOutputManager operations
func BenchmarkAudioOutputManager(b *testing.B) {
b.Run("Start", func(b *testing.B) {
for i := 0; i < b.N; i++ {
manager := NewAudioOutputManager()
manager.Start()
manager.Stop()
}
})
b.Run("IsRunning", func(b *testing.B) {
manager := NewAudioOutputManager()
manager.Start()
defer manager.Stop()
b.ResetTimer()
for i := 0; i < b.N; i++ {
manager.IsRunning()
}
})
b.Run("GetMetrics", func(b *testing.B) {
manager := NewAudioOutputManager()
manager.Start()
defer manager.Stop()
b.ResetTimer()
for i := 0; i < b.N; i++ {
manager.GetMetrics()
}
})
}

View File

@ -17,7 +17,7 @@ func RunAudioOutputServer() error {
logger.Info().Msg("Starting audio output server subprocess")
// Create audio server
server, err := NewAudioServer()
server, err := NewAudioOutputServer()
if err != nil {
logger.Error().Err(err).Msg("failed to create audio server")
return err

View File

@ -12,23 +12,24 @@ import (
"github.com/rs/zerolog"
)
// OutputStreamer manages high-performance audio output streaming
type OutputStreamer struct {
// Atomic fields must be first for proper alignment on ARM
// AudioOutputStreamer manages high-performance audio output streaming
type AudioOutputStreamer struct {
// Performance metrics (atomic operations for thread safety)
processedFrames int64 // Total processed frames counter (atomic)
droppedFrames int64 // Dropped frames counter (atomic)
processingTime int64 // Average processing time in nanoseconds (atomic)
lastStatsTime int64 // Last statistics update time (atomic)
client *AudioClient
client *AudioOutputClient
bufferPool *AudioBufferPool
ctx context.Context
cancel context.CancelFunc
wg sync.WaitGroup
running bool
mtx sync.Mutex
chanClosed bool // Track if processing channel is closed
// Performance optimization fields
// Adaptive processing configuration
batchSize int // Adaptive batch size for frame processing
processingChan chan []byte // Buffered channel for frame processing
statsInterval time.Duration // Statistics reporting interval
@ -42,21 +43,21 @@ var (
func getOutputStreamingLogger() *zerolog.Logger {
if outputStreamingLogger == nil {
logger := logging.GetDefaultLogger().With().Str("component", "audio-output").Logger()
logger := logging.GetDefaultLogger().With().Str("component", AudioOutputStreamerComponent).Logger()
outputStreamingLogger = &logger
}
return outputStreamingLogger
}
func NewOutputStreamer() (*OutputStreamer, error) {
client := NewAudioClient()
func NewAudioOutputStreamer() (*AudioOutputStreamer, error) {
client := NewAudioOutputClient()
// Get initial batch size from adaptive buffer manager
adaptiveManager := GetAdaptiveBufferManager()
initialBatchSize := adaptiveManager.GetOutputBufferSize()
ctx, cancel := context.WithCancel(context.Background())
return &OutputStreamer{
return &AudioOutputStreamer{
client: client,
bufferPool: NewAudioBufferPool(GetMaxAudioFrameSize()), // Use existing buffer pool
ctx: ctx,
@ -68,7 +69,7 @@ func NewOutputStreamer() (*OutputStreamer, error) {
}, nil
}
func (s *OutputStreamer) Start() error {
func (s *AudioOutputStreamer) Start() error {
s.mtx.Lock()
defer s.mtx.Unlock()
@ -92,7 +93,7 @@ func (s *OutputStreamer) Start() error {
return nil
}
func (s *OutputStreamer) Stop() {
func (s *AudioOutputStreamer) Stop() {
s.mtx.Lock()
defer s.mtx.Unlock()
@ -103,8 +104,11 @@ func (s *OutputStreamer) Stop() {
s.running = false
s.cancel()
// Close processing channel to signal goroutines
close(s.processingChan)
// Close processing channel to signal goroutines (only if not already closed)
if !s.chanClosed {
close(s.processingChan)
s.chanClosed = true
}
// Wait for all goroutines to finish
s.wg.Wait()
@ -114,7 +118,7 @@ func (s *OutputStreamer) Stop() {
}
}
func (s *OutputStreamer) streamLoop() {
func (s *AudioOutputStreamer) streamLoop() {
defer s.wg.Done()
// Pin goroutine to OS thread for consistent performance
@ -153,7 +157,9 @@ func (s *OutputStreamer) streamLoop() {
if n > 0 {
// Send frame for processing (non-blocking)
frameData := make([]byte, n)
// Use buffer pool to avoid allocation
frameData := s.bufferPool.Get()
frameData = frameData[:n]
copy(frameData, frameBuf[:n])
select {
@ -175,7 +181,7 @@ func (s *OutputStreamer) streamLoop() {
}
// processingLoop handles frame processing in a separate goroutine
func (s *OutputStreamer) processingLoop() {
func (s *AudioOutputStreamer) processingLoop() {
defer s.wg.Done()
// Pin goroutine to OS thread for consistent performance
@ -192,25 +198,29 @@ func (s *OutputStreamer) processingLoop() {
}
}()
for range s.processingChan {
// Process frame (currently just receiving, but can be extended)
if _, err := s.client.ReceiveFrame(); err != nil {
if s.client.IsConnected() {
getOutputStreamingLogger().Warn().Err(err).Msg("Error reading audio frame from output server")
atomic.AddInt64(&s.droppedFrames, 1)
}
// Try to reconnect if disconnected
if !s.client.IsConnected() {
if err := s.client.Connect(); err != nil {
getOutputStreamingLogger().Warn().Err(err).Msg("Failed to reconnect")
for frameData := range s.processingChan {
// Process frame and return buffer to pool after processing
func() {
defer s.bufferPool.Put(frameData)
if _, err := s.client.ReceiveFrame(); err != nil {
if s.client.IsConnected() {
getOutputStreamingLogger().Warn().Err(err).Msg("Error reading audio frame from output server")
atomic.AddInt64(&s.droppedFrames, 1)
}
// Try to reconnect if disconnected
if !s.client.IsConnected() {
if err := s.client.Connect(); err != nil {
getOutputStreamingLogger().Warn().Err(err).Msg("Failed to reconnect")
}
}
}
}
}()
}
}
// statisticsLoop monitors and reports performance statistics
func (s *OutputStreamer) statisticsLoop() {
func (s *AudioOutputStreamer) statisticsLoop() {
defer s.wg.Done()
ticker := time.NewTicker(s.statsInterval)
@ -227,7 +237,7 @@ func (s *OutputStreamer) statisticsLoop() {
}
// reportStatistics logs current performance statistics
func (s *OutputStreamer) reportStatistics() {
func (s *AudioOutputStreamer) reportStatistics() {
processed := atomic.LoadInt64(&s.processedFrames)
dropped := atomic.LoadInt64(&s.droppedFrames)
processingTime := atomic.LoadInt64(&s.processingTime)
@ -245,7 +255,7 @@ func (s *OutputStreamer) reportStatistics() {
}
// GetStats returns streaming statistics
func (s *OutputStreamer) GetStats() (processed, dropped int64, avgProcessingTime time.Duration) {
func (s *AudioOutputStreamer) GetStats() (processed, dropped int64, avgProcessingTime time.Duration) {
processed = atomic.LoadInt64(&s.processedFrames)
dropped = atomic.LoadInt64(&s.droppedFrames)
processingTimeNs := atomic.LoadInt64(&s.processingTime)
@ -254,7 +264,7 @@ func (s *OutputStreamer) GetStats() (processed, dropped int64, avgProcessingTime
}
// GetDetailedStats returns comprehensive streaming statistics
func (s *OutputStreamer) GetDetailedStats() map[string]interface{} {
func (s *AudioOutputStreamer) GetDetailedStats() map[string]interface{} {
processed := atomic.LoadInt64(&s.processedFrames)
dropped := atomic.LoadInt64(&s.droppedFrames)
processingTime := atomic.LoadInt64(&s.processingTime)
@ -282,7 +292,7 @@ func (s *OutputStreamer) GetDetailedStats() map[string]interface{} {
}
// UpdateBatchSize updates the batch size from adaptive buffer manager
func (s *OutputStreamer) UpdateBatchSize() {
func (s *AudioOutputStreamer) UpdateBatchSize() {
s.mtx.Lock()
adaptiveManager := GetAdaptiveBufferManager()
s.batchSize = adaptiveManager.GetOutputBufferSize()
@ -290,7 +300,7 @@ func (s *OutputStreamer) UpdateBatchSize() {
}
// ReportLatency reports processing latency to adaptive buffer manager
func (s *OutputStreamer) ReportLatency(latency time.Duration) {
func (s *AudioOutputStreamer) ReportLatency(latency time.Duration) {
adaptiveManager := GetAdaptiveBufferManager()
adaptiveManager.UpdateLatency(latency)
}
@ -321,17 +331,61 @@ func StartAudioOutputStreaming(send func([]byte)) error {
getOutputStreamingLogger().Info().Str("socket_path", getOutputSocketPath()).Msg("Audio output streaming started, connected to output server")
buffer := make([]byte, GetMaxAudioFrameSize())
consecutiveErrors := 0
maxConsecutiveErrors := GetConfig().MaxConsecutiveErrors
errorBackoffDelay := GetConfig().RetryDelay
maxErrorBackoff := GetConfig().MaxRetryDelay
for {
select {
case <-ctx.Done():
return
default:
// Capture audio frame
// Capture audio frame with enhanced error handling
n, err := CGOAudioReadEncode(buffer)
if err != nil {
getOutputStreamingLogger().Warn().Err(err).Msg("Failed to read/encode audio")
consecutiveErrors++
getOutputStreamingLogger().Warn().
Err(err).
Int("consecutive_errors", consecutiveErrors).
Msg("Failed to read/encode audio")
// Implement progressive backoff for consecutive errors
if consecutiveErrors >= maxConsecutiveErrors {
getOutputStreamingLogger().Error().
Int("consecutive_errors", consecutiveErrors).
Msg("Too many consecutive audio errors, attempting recovery")
// Try to reinitialize audio system
CGOAudioClose()
time.Sleep(errorBackoffDelay)
if initErr := CGOAudioInit(); initErr != nil {
getOutputStreamingLogger().Error().
Err(initErr).
Msg("Failed to reinitialize audio system")
// Exponential backoff for reinitialization failures
errorBackoffDelay = time.Duration(float64(errorBackoffDelay) * GetConfig().BackoffMultiplier)
if errorBackoffDelay > maxErrorBackoff {
errorBackoffDelay = maxErrorBackoff
}
} else {
getOutputStreamingLogger().Info().Msg("Audio system reinitialized successfully")
consecutiveErrors = 0
errorBackoffDelay = GetConfig().RetryDelay // Reset backoff
}
} else {
// Brief delay for transient errors
time.Sleep(GetConfig().ShortSleepDuration)
}
continue
}
// Success - reset error counters
if consecutiveErrors > 0 {
consecutiveErrors = 0
errorBackoffDelay = GetConfig().RetryDelay
}
if n > 0 {
// Get frame buffer from pool to reduce allocations
frame := GetAudioFrameBuffer()

View File

@ -0,0 +1,341 @@
package audio
import (
"sync"
"testing"
"time"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
// TestAudioOutputStreamer tests the AudioOutputStreamer component
func TestAudioOutputStreamer(t *testing.T) {
tests := []struct {
name string
testFunc func(t *testing.T)
}{
{"NewAudioOutputStreamer", testNewAudioOutputStreamer},
{"Start", testAudioOutputStreamerStart},
{"Stop", testAudioOutputStreamerStop},
{"StartStop", testAudioOutputStreamerStartStop},
{"GetStats", testAudioOutputStreamerGetStats},
{"GetDetailedStats", testAudioOutputStreamerGetDetailedStats},
{"UpdateBatchSize", testAudioOutputStreamerUpdateBatchSize},
{"ReportLatency", testAudioOutputStreamerReportLatency},
{"ConcurrentOperations", testAudioOutputStreamerConcurrent},
{"MultipleStarts", testAudioOutputStreamerMultipleStarts},
{"MultipleStops", testAudioOutputStreamerMultipleStops},
}
for _, tt := range tests {
t.Run(tt.name, func(t *testing.T) {
tt.testFunc(t)
})
}
}
func testNewAudioOutputStreamer(t *testing.T) {
streamer, err := NewAudioOutputStreamer()
if err != nil {
// If creation fails due to missing dependencies, skip the test
t.Skipf("Skipping test due to missing dependencies: %v", err)
return
}
require.NotNil(t, streamer)
// Test initial state
processed, dropped, avgTime := streamer.GetStats()
assert.GreaterOrEqual(t, processed, int64(0))
assert.GreaterOrEqual(t, dropped, int64(0))
assert.GreaterOrEqual(t, avgTime, time.Duration(0))
// Cleanup
streamer.Stop()
}
func testAudioOutputStreamerStart(t *testing.T) {
streamer, err := NewAudioOutputStreamer()
if err != nil {
t.Skipf("Skipping test due to missing dependencies: %v", err)
return
}
require.NotNil(t, streamer)
// Test start
err = streamer.Start()
assert.NoError(t, err)
// Cleanup
streamer.Stop()
}
func testAudioOutputStreamerStop(t *testing.T) {
streamer, err := NewAudioOutputStreamer()
if err != nil {
t.Skipf("Skipping test due to missing dependencies: %v", err)
return
}
require.NotNil(t, streamer)
// Start first
err = streamer.Start()
require.NoError(t, err)
// Test stop
streamer.Stop()
// Multiple stops should be safe
streamer.Stop()
streamer.Stop()
}
func testAudioOutputStreamerStartStop(t *testing.T) {
streamer, err := NewAudioOutputStreamer()
if err != nil {
t.Skipf("Skipping test due to missing dependencies: %v", err)
return
}
require.NotNil(t, streamer)
// Test multiple start/stop cycles
for i := 0; i < 3; i++ {
// Start
err = streamer.Start()
assert.NoError(t, err)
// Stop
streamer.Stop()
}
}
func testAudioOutputStreamerGetStats(t *testing.T) {
streamer, err := NewAudioOutputStreamer()
if err != nil {
t.Skipf("Skipping test due to missing dependencies: %v", err)
return
}
require.NotNil(t, streamer)
// Test stats when not running
processed, dropped, avgTime := streamer.GetStats()
assert.Equal(t, int64(0), processed)
assert.Equal(t, int64(0), dropped)
assert.GreaterOrEqual(t, avgTime, time.Duration(0))
// Start and test stats
err = streamer.Start()
require.NoError(t, err)
processed, dropped, avgTime = streamer.GetStats()
assert.GreaterOrEqual(t, processed, int64(0))
assert.GreaterOrEqual(t, dropped, int64(0))
assert.GreaterOrEqual(t, avgTime, time.Duration(0))
// Cleanup
streamer.Stop()
}
func testAudioOutputStreamerGetDetailedStats(t *testing.T) {
streamer, err := NewAudioOutputStreamer()
if err != nil {
t.Skipf("Skipping test due to missing dependencies: %v", err)
return
}
require.NotNil(t, streamer)
// Test detailed stats
stats := streamer.GetDetailedStats()
assert.NotNil(t, stats)
assert.Contains(t, stats, "processed_frames")
assert.Contains(t, stats, "dropped_frames")
assert.Contains(t, stats, "batch_size")
assert.Contains(t, stats, "connected")
assert.Equal(t, int64(0), stats["processed_frames"])
assert.Equal(t, int64(0), stats["dropped_frames"])
// Start and test detailed stats
err = streamer.Start()
require.NoError(t, err)
stats = streamer.GetDetailedStats()
assert.NotNil(t, stats)
assert.Contains(t, stats, "processed_frames")
assert.Contains(t, stats, "dropped_frames")
// Cleanup
streamer.Stop()
}
func testAudioOutputStreamerUpdateBatchSize(t *testing.T) {
streamer, err := NewAudioOutputStreamer()
if err != nil {
t.Skipf("Skipping test due to missing dependencies: %v", err)
return
}
require.NotNil(t, streamer)
// Test updating batch size (no parameters, uses adaptive manager)
streamer.UpdateBatchSize()
streamer.UpdateBatchSize()
streamer.UpdateBatchSize()
// Cleanup
streamer.Stop()
}
func testAudioOutputStreamerReportLatency(t *testing.T) {
streamer, err := NewAudioOutputStreamer()
if err != nil {
t.Skipf("Skipping test due to missing dependencies: %v", err)
return
}
require.NotNil(t, streamer)
// Test reporting latency
streamer.ReportLatency(10 * time.Millisecond)
streamer.ReportLatency(5 * time.Millisecond)
streamer.ReportLatency(15 * time.Millisecond)
// Cleanup
streamer.Stop()
}
func testAudioOutputStreamerConcurrent(t *testing.T) {
streamer, err := NewAudioOutputStreamer()
if err != nil {
t.Skipf("Skipping test due to missing dependencies: %v", err)
return
}
require.NotNil(t, streamer)
var wg sync.WaitGroup
const numGoroutines = 10
// Test concurrent starts
wg.Add(numGoroutines)
for i := 0; i < numGoroutines; i++ {
go func() {
defer wg.Done()
streamer.Start()
}()
}
wg.Wait()
// Test concurrent operations
wg.Add(numGoroutines * 3)
for i := 0; i < numGoroutines; i++ {
go func() {
defer wg.Done()
streamer.GetStats()
}()
go func() {
defer wg.Done()
streamer.UpdateBatchSize()
}()
go func() {
defer wg.Done()
streamer.ReportLatency(10 * time.Millisecond)
}()
}
wg.Wait()
// Test concurrent stops
wg.Add(numGoroutines)
for i := 0; i < numGoroutines; i++ {
go func() {
defer wg.Done()
streamer.Stop()
}()
}
wg.Wait()
}
func testAudioOutputStreamerMultipleStarts(t *testing.T) {
streamer, err := NewAudioOutputStreamer()
if err != nil {
t.Skipf("Skipping test due to missing dependencies: %v", err)
return
}
require.NotNil(t, streamer)
// First start should succeed
err = streamer.Start()
assert.NoError(t, err)
// Subsequent starts should return error
err = streamer.Start()
assert.Error(t, err)
assert.Contains(t, err.Error(), "already running")
err = streamer.Start()
assert.Error(t, err)
assert.Contains(t, err.Error(), "already running")
// Cleanup
streamer.Stop()
}
func testAudioOutputStreamerMultipleStops(t *testing.T) {
streamer, err := NewAudioOutputStreamer()
if err != nil {
t.Skipf("Skipping test due to missing dependencies: %v", err)
return
}
require.NotNil(t, streamer)
// Start first
err = streamer.Start()
require.NoError(t, err)
// Multiple stops should be safe
streamer.Stop()
streamer.Stop()
streamer.Stop()
}
// BenchmarkAudioOutputStreamer benchmarks the AudioOutputStreamer operations
func BenchmarkAudioOutputStreamer(b *testing.B) {
b.Run("GetStats", func(b *testing.B) {
streamer, err := NewAudioOutputStreamer()
if err != nil {
b.Skipf("Skipping benchmark due to missing dependencies: %v", err)
return
}
defer streamer.Stop()
streamer.Start()
b.ResetTimer()
for i := 0; i < b.N; i++ {
streamer.GetStats()
}
})
b.Run("UpdateBatchSize", func(b *testing.B) {
streamer, err := NewAudioOutputStreamer()
if err != nil {
b.Skipf("Skipping benchmark due to missing dependencies: %v", err)
return
}
defer streamer.Stop()
b.ResetTimer()
for i := 0; i < b.N; i++ {
streamer.UpdateBatchSize()
}
})
b.Run("ReportLatency", func(b *testing.B) {
streamer, err := NewAudioOutputStreamer()
if err != nil {
b.Skipf("Skipping benchmark due to missing dependencies: %v", err)
return
}
defer streamer.Stop()
b.ResetTimer()
for i := 0; i < b.N; i++ {
streamer.ReportLatency(10 * time.Millisecond)
}
})
}

View File

@ -19,13 +19,14 @@ type AudioRelay struct {
framesRelayed int64
framesDropped int64
client *AudioClient
ctx context.Context
cancel context.CancelFunc
wg sync.WaitGroup
logger *zerolog.Logger
running bool
mutex sync.RWMutex
client *AudioOutputClient
ctx context.Context
cancel context.CancelFunc
wg sync.WaitGroup
logger *zerolog.Logger
running bool
mutex sync.RWMutex
bufferPool *AudioBufferPool // Buffer pool for memory optimization
// WebRTC integration
audioTrack AudioTrackWriter
@ -44,9 +45,10 @@ func NewAudioRelay() *AudioRelay {
logger := logging.GetDefaultLogger().With().Str("component", "audio-relay").Logger()
return &AudioRelay{
ctx: ctx,
cancel: cancel,
logger: &logger,
ctx: ctx,
cancel: cancel,
logger: &logger,
bufferPool: NewAudioBufferPool(GetMaxAudioFrameSize()),
}
}
@ -60,7 +62,7 @@ func (r *AudioRelay) Start(audioTrack AudioTrackWriter, config AudioConfig) erro
}
// Create audio client to connect to subprocess
client := NewAudioClient()
client := NewAudioOutputClient()
r.client = client
r.audioTrack = audioTrack
r.config = config
@ -188,8 +190,14 @@ func (r *AudioRelay) forwardToWebRTC(frame []byte) error {
// Prepare sample data
var sampleData []byte
if muted {
// Send silence when muted
sampleData = make([]byte, len(frame))
// Send silence when muted - use buffer pool to avoid allocation
sampleData = r.bufferPool.Get()
sampleData = sampleData[:len(frame)] // Resize to frame length
// Clear the buffer to create silence
for i := range sampleData {
sampleData[i] = 0
}
defer r.bufferPool.Put(sampleData) // Return to pool after use
} else {
sampleData = frame
}

View File

@ -34,8 +34,8 @@ func getMaxRestartDelay() time.Duration {
return GetConfig().MaxRestartDelay
}
// AudioServerSupervisor manages the audio server subprocess lifecycle
type AudioServerSupervisor struct {
// AudioOutputSupervisor manages the audio output server subprocess lifecycle
type AudioOutputSupervisor struct {
ctx context.Context
cancel context.CancelFunc
logger *zerolog.Logger
@ -52,8 +52,10 @@ type AudioServerSupervisor struct {
lastExitTime time.Time
// Channels for coordination
processDone chan struct{}
stopChan chan struct{}
processDone chan struct{}
stopChan chan struct{}
stopChanClosed bool // Track if stopChan is closed
processDoneClosed bool // Track if processDone is closed
// Process monitoring
processMonitor *ProcessMonitor
@ -64,12 +66,12 @@ type AudioServerSupervisor struct {
onRestart func(attempt int, delay time.Duration)
}
// NewAudioServerSupervisor creates a new audio server supervisor
func NewAudioServerSupervisor() *AudioServerSupervisor {
// NewAudioOutputSupervisor creates a new audio output server supervisor
func NewAudioOutputSupervisor() *AudioOutputSupervisor {
ctx, cancel := context.WithCancel(context.Background())
logger := logging.GetDefaultLogger().With().Str("component", "audio-supervisor").Logger()
logger := logging.GetDefaultLogger().With().Str("component", AudioOutputSupervisorComponent).Logger()
return &AudioServerSupervisor{
return &AudioOutputSupervisor{
ctx: ctx,
cancel: cancel,
logger: &logger,
@ -80,7 +82,7 @@ func NewAudioServerSupervisor() *AudioServerSupervisor {
}
// SetCallbacks sets optional callbacks for process lifecycle events
func (s *AudioServerSupervisor) SetCallbacks(
func (s *AudioOutputSupervisor) SetCallbacks(
onStart func(pid int),
onExit func(pid int, exitCode int, crashed bool),
onRestart func(attempt int, delay time.Duration),
@ -93,79 +95,100 @@ func (s *AudioServerSupervisor) SetCallbacks(
s.onRestart = onRestart
}
// Start begins supervising the audio server process
func (s *AudioServerSupervisor) Start() error {
// Start begins supervising the audio output server process
func (s *AudioOutputSupervisor) Start() error {
if !atomic.CompareAndSwapInt32(&s.running, 0, 1) {
return fmt.Errorf("supervisor already running")
return fmt.Errorf("audio output supervisor is already running")
}
s.logger.Info().Msg("starting audio server supervisor")
s.logger.Info().Str("component", AudioOutputSupervisorComponent).Msg("starting component")
// Recreate channels in case they were closed by a previous Stop() call
s.mutex.Lock()
s.processDone = make(chan struct{})
s.stopChan = make(chan struct{})
s.stopChanClosed = false // Reset channel closed flag
s.processDoneClosed = false // Reset channel closed flag
// Recreate context as well since it might have been cancelled
s.ctx, s.cancel = context.WithCancel(context.Background())
// Reset restart tracking on start
s.restartAttempts = s.restartAttempts[:0]
s.lastExitCode = 0
s.lastExitTime = time.Time{}
s.mutex.Unlock()
// Start the supervision loop
go s.supervisionLoop()
s.logger.Info().Str("component", AudioOutputSupervisorComponent).Msg("component started successfully")
return nil
}
// Stop gracefully stops the audio server and supervisor
func (s *AudioServerSupervisor) Stop() error {
func (s *AudioOutputSupervisor) Stop() {
if !atomic.CompareAndSwapInt32(&s.running, 1, 0) {
return nil // Already stopped
return // Already stopped
}
s.logger.Info().Msg("stopping audio server supervisor")
s.logger.Info().Str("component", AudioOutputSupervisorComponent).Msg("stopping component")
// Signal stop and wait for cleanup
close(s.stopChan)
s.mutex.Lock()
if !s.stopChanClosed {
close(s.stopChan)
s.stopChanClosed = true
}
s.mutex.Unlock()
s.cancel()
// Wait for process to exit
select {
case <-s.processDone:
s.logger.Info().Msg("audio server process stopped gracefully")
s.logger.Info().Str("component", AudioOutputSupervisorComponent).Msg("component stopped gracefully")
case <-time.After(GetConfig().SupervisorTimeout):
s.logger.Warn().Msg("audio server process did not stop gracefully, forcing termination")
s.logger.Warn().Str("component", AudioOutputSupervisorComponent).Msg("component did not stop gracefully, forcing termination")
s.forceKillProcess()
}
return nil
s.logger.Info().Str("component", AudioOutputSupervisorComponent).Msg("component stopped")
}
// IsRunning returns true if the supervisor is running
func (s *AudioServerSupervisor) IsRunning() bool {
func (s *AudioOutputSupervisor) IsRunning() bool {
return atomic.LoadInt32(&s.running) == 1
}
// GetProcessPID returns the current process PID (0 if not running)
func (s *AudioServerSupervisor) GetProcessPID() int {
func (s *AudioOutputSupervisor) GetProcessPID() int {
s.mutex.RLock()
defer s.mutex.RUnlock()
return s.processPID
}
// GetLastExitInfo returns information about the last process exit
func (s *AudioServerSupervisor) GetLastExitInfo() (exitCode int, exitTime time.Time) {
func (s *AudioOutputSupervisor) GetLastExitInfo() (exitCode int, exitTime time.Time) {
s.mutex.RLock()
defer s.mutex.RUnlock()
return s.lastExitCode, s.lastExitTime
}
// GetProcessMetrics returns current process metrics if the process is running
func (s *AudioServerSupervisor) GetProcessMetrics() *ProcessMetrics {
func (s *AudioOutputSupervisor) GetProcessMetrics() *ProcessMetrics {
s.mutex.RLock()
pid := s.processPID
s.mutex.RUnlock()
if pid == 0 {
return nil
// Return default metrics when no process is running
return &ProcessMetrics{
PID: 0,
CPUPercent: 0.0,
MemoryRSS: 0,
MemoryVMS: 0,
MemoryPercent: 0.0,
Timestamp: time.Now(),
ProcessName: "audio-output-server",
}
}
metrics := s.processMonitor.GetCurrentMetrics()
@ -174,13 +197,28 @@ func (s *AudioServerSupervisor) GetProcessMetrics() *ProcessMetrics {
return &metric
}
}
return nil
// Return default metrics if process not found in monitor
return &ProcessMetrics{
PID: pid,
CPUPercent: 0.0,
MemoryRSS: 0,
MemoryVMS: 0,
MemoryPercent: 0.0,
Timestamp: time.Now(),
ProcessName: "audio-output-server",
}
}
// supervisionLoop is the main supervision loop
func (s *AudioServerSupervisor) supervisionLoop() {
func (s *AudioOutputSupervisor) supervisionLoop() {
defer func() {
close(s.processDone)
s.mutex.Lock()
if !s.processDoneClosed {
close(s.processDone)
s.processDoneClosed = true
}
s.mutex.Unlock()
s.logger.Info().Msg("audio server supervision ended")
}()
@ -252,7 +290,7 @@ func (s *AudioServerSupervisor) supervisionLoop() {
}
// startProcess starts the audio server process
func (s *AudioServerSupervisor) startProcess() error {
func (s *AudioOutputSupervisor) startProcess() error {
execPath, err := os.Executable()
if err != nil {
return fmt.Errorf("failed to get executable path: %w", err)
@ -285,7 +323,7 @@ func (s *AudioServerSupervisor) startProcess() error {
}
// waitForProcessExit waits for the current process to exit and logs the result
func (s *AudioServerSupervisor) waitForProcessExit() {
func (s *AudioOutputSupervisor) waitForProcessExit() {
s.mutex.RLock()
cmd := s.cmd
pid := s.processPID
@ -338,7 +376,7 @@ func (s *AudioServerSupervisor) waitForProcessExit() {
}
// terminateProcess gracefully terminates the current process
func (s *AudioServerSupervisor) terminateProcess() {
func (s *AudioOutputSupervisor) terminateProcess() {
s.mutex.RLock()
cmd := s.cmd
pid := s.processPID
@ -365,14 +403,14 @@ func (s *AudioServerSupervisor) terminateProcess() {
select {
case <-done:
s.logger.Info().Int("pid", pid).Msg("audio server process terminated gracefully")
case <-time.After(GetConfig().InputSupervisorTimeout):
case <-time.After(GetConfig().OutputSupervisorTimeout):
s.logger.Warn().Int("pid", pid).Msg("process did not terminate gracefully, sending SIGKILL")
s.forceKillProcess()
}
}
// forceKillProcess forcefully kills the current process
func (s *AudioServerSupervisor) forceKillProcess() {
func (s *AudioOutputSupervisor) forceKillProcess() {
s.mutex.RLock()
cmd := s.cmd
pid := s.processPID
@ -389,7 +427,7 @@ func (s *AudioServerSupervisor) forceKillProcess() {
}
// shouldRestart determines if the process should be restarted
func (s *AudioServerSupervisor) shouldRestart() bool {
func (s *AudioOutputSupervisor) shouldRestart() bool {
if atomic.LoadInt32(&s.running) == 0 {
return false // Supervisor is stopping
}
@ -411,7 +449,7 @@ func (s *AudioServerSupervisor) shouldRestart() bool {
}
// recordRestartAttempt records a restart attempt
func (s *AudioServerSupervisor) recordRestartAttempt() {
func (s *AudioOutputSupervisor) recordRestartAttempt() {
s.mutex.Lock()
defer s.mutex.Unlock()
@ -419,7 +457,7 @@ func (s *AudioServerSupervisor) recordRestartAttempt() {
}
// calculateRestartDelay calculates the delay before next restart attempt
func (s *AudioServerSupervisor) calculateRestartDelay() time.Duration {
func (s *AudioOutputSupervisor) calculateRestartDelay() time.Duration {
s.mutex.RLock()
defer s.mutex.RUnlock()

View File

@ -0,0 +1,217 @@
package audio
import (
"sync"
"testing"
"time"
"github.com/stretchr/testify/assert"
"github.com/stretchr/testify/require"
)
func TestNewAudioOutputSupervisor(t *testing.T) {
supervisor := NewAudioOutputSupervisor()
assert.NotNil(t, supervisor)
assert.False(t, supervisor.IsRunning())
}
func TestAudioOutputSupervisorStart(t *testing.T) {
supervisor := NewAudioOutputSupervisor()
require.NotNil(t, supervisor)
// Test successful start
err := supervisor.Start()
assert.NoError(t, err)
assert.True(t, supervisor.IsRunning())
// Test starting already running supervisor
err = supervisor.Start()
assert.Error(t, err)
assert.Contains(t, err.Error(), "already running")
// Cleanup
supervisor.Stop()
}
func TestAudioOutputSupervisorStop(t *testing.T) {
supervisor := NewAudioOutputSupervisor()
require.NotNil(t, supervisor)
// Test stopping non-running supervisor
supervisor.Stop()
assert.False(t, supervisor.IsRunning())
// Start and then stop
err := supervisor.Start()
require.NoError(t, err)
assert.True(t, supervisor.IsRunning())
supervisor.Stop()
assert.False(t, supervisor.IsRunning())
}
func TestAudioOutputSupervisorIsRunning(t *testing.T) {
supervisor := NewAudioOutputSupervisor()
require.NotNil(t, supervisor)
// Test initial state
assert.False(t, supervisor.IsRunning())
// Test after start
err := supervisor.Start()
require.NoError(t, err)
assert.True(t, supervisor.IsRunning())
// Test after stop
supervisor.Stop()
assert.False(t, supervisor.IsRunning())
}
func TestAudioOutputSupervisorGetProcessMetrics(t *testing.T) {
supervisor := NewAudioOutputSupervisor()
require.NotNil(t, supervisor)
// Test metrics when not running
metrics := supervisor.GetProcessMetrics()
assert.NotNil(t, metrics)
// Start and test metrics
err := supervisor.Start()
require.NoError(t, err)
metrics = supervisor.GetProcessMetrics()
assert.NotNil(t, metrics)
// Cleanup
supervisor.Stop()
}
func TestAudioOutputSupervisorConcurrentOperations(t *testing.T) {
supervisor := NewAudioOutputSupervisor()
require.NotNil(t, supervisor)
var wg sync.WaitGroup
// Test concurrent start/stop operations
for i := 0; i < 10; i++ {
wg.Add(2)
go func() {
defer wg.Done()
_ = supervisor.Start()
}()
go func() {
defer wg.Done()
supervisor.Stop()
}()
}
// Test concurrent metric access
for i := 0; i < 5; i++ {
wg.Add(1)
go func() {
defer wg.Done()
_ = supervisor.GetProcessMetrics()
}()
}
// Test concurrent status checks
for i := 0; i < 5; i++ {
wg.Add(1)
go func() {
defer wg.Done()
_ = supervisor.IsRunning()
}()
}
wg.Wait()
// Cleanup
supervisor.Stop()
}
func TestAudioOutputSupervisorMultipleStartStop(t *testing.T) {
supervisor := NewAudioOutputSupervisor()
require.NotNil(t, supervisor)
// Test multiple start/stop cycles
for i := 0; i < 5; i++ {
err := supervisor.Start()
assert.NoError(t, err)
assert.True(t, supervisor.IsRunning())
supervisor.Stop()
assert.False(t, supervisor.IsRunning())
}
}
func TestAudioOutputSupervisorHealthCheck(t *testing.T) {
supervisor := NewAudioOutputSupervisor()
require.NotNil(t, supervisor)
// Start supervisor
err := supervisor.Start()
require.NoError(t, err)
// Give some time for health monitoring to initialize
time.Sleep(100 * time.Millisecond)
// Test that supervisor is still running
assert.True(t, supervisor.IsRunning())
// Cleanup
supervisor.Stop()
}
func TestAudioOutputSupervisorProcessManagement(t *testing.T) {
supervisor := NewAudioOutputSupervisor()
require.NotNil(t, supervisor)
// Start supervisor
err := supervisor.Start()
require.NoError(t, err)
// Give some time for process management to initialize
time.Sleep(200 * time.Millisecond)
// Test that supervisor is managing processes
assert.True(t, supervisor.IsRunning())
// Cleanup
supervisor.Stop()
// Ensure supervisor stopped cleanly
assert.False(t, supervisor.IsRunning())
}
// Benchmark tests
func BenchmarkAudioOutputSupervisor(b *testing.B) {
supervisor := NewAudioOutputSupervisor()
b.Run("Start", func(b *testing.B) {
b.ResetTimer()
for i := 0; i < b.N; i++ {
_ = supervisor.Start()
supervisor.Stop()
}
})
b.Run("GetProcessMetrics", func(b *testing.B) {
_ = supervisor.Start()
defer supervisor.Stop()
b.ResetTimer()
for i := 0; i < b.N; i++ {
_ = supervisor.GetProcessMetrics()
}
})
b.Run("IsRunning", func(b *testing.B) {
_ = supervisor.Start()
defer supervisor.Stop()
b.ResetTimer()
for i := 0; i < b.N; i++ {
_ = supervisor.IsRunning()
}
})
}

View File

@ -0,0 +1,177 @@
package audio
import (
"errors"
"time"
)
// Validation errors
var (
ErrInvalidAudioQuality = errors.New("invalid audio quality level")
ErrInvalidFrameSize = errors.New("invalid frame size")
ErrInvalidFrameData = errors.New("invalid frame data")
ErrInvalidBufferSize = errors.New("invalid buffer size")
ErrInvalidPriority = errors.New("invalid priority value")
ErrInvalidLatency = errors.New("invalid latency value")
ErrInvalidConfiguration = errors.New("invalid configuration")
ErrInvalidSocketConfig = errors.New("invalid socket configuration")
ErrInvalidMetricsInterval = errors.New("invalid metrics interval")
ErrInvalidSampleRate = errors.New("invalid sample rate")
ErrInvalidChannels = errors.New("invalid channels")
)
// ValidateAudioQuality validates audio quality enum values
func ValidateAudioQuality(quality AudioQuality) error {
switch quality {
case AudioQualityLow, AudioQualityMedium, AudioQualityHigh, AudioQualityUltra:
return nil
default:
return ErrInvalidAudioQuality
}
}
// ValidateFrameData validates audio frame data
func ValidateFrameData(data []byte) error {
if len(data) == 0 {
return ErrInvalidFrameData
}
// Use a reasonable default if config is not available
maxFrameSize := 4096
if config := GetConfig(); config != nil {
maxFrameSize = config.MaxAudioFrameSize
}
if len(data) > maxFrameSize {
return ErrInvalidFrameSize
}
return nil
}
// ValidateZeroCopyFrame validates zero-copy audio frame
func ValidateZeroCopyFrame(frame *ZeroCopyAudioFrame) error {
if frame == nil {
return ErrInvalidFrameData
}
data := frame.Data()
if len(data) == 0 {
return ErrInvalidFrameData
}
// Use a reasonable default if config is not available
maxFrameSize := 4096
if config := GetConfig(); config != nil {
maxFrameSize = config.MaxAudioFrameSize
}
if len(data) > maxFrameSize {
return ErrInvalidFrameSize
}
return nil
}
// ValidateBufferSize validates buffer size parameters
func ValidateBufferSize(size int) error {
if size <= 0 {
return ErrInvalidBufferSize
}
// Use a reasonable default if config is not available
maxBuffer := 262144 // 256KB default
if config := GetConfig(); config != nil {
maxBuffer = config.SocketMaxBuffer
}
if size > maxBuffer {
return ErrInvalidBufferSize
}
return nil
}
// ValidateThreadPriority validates thread priority values
func ValidateThreadPriority(priority int) error {
// Use reasonable defaults if config is not available
minPriority := -20
maxPriority := 99
if config := GetConfig(); config != nil {
minPriority = config.MinNiceValue
maxPriority = config.RTAudioHighPriority
}
if priority < minPriority || priority > maxPriority {
return ErrInvalidPriority
}
return nil
}
// ValidateLatency validates latency values
func ValidateLatency(latency time.Duration) error {
if latency < 0 {
return ErrInvalidLatency
}
// Use a reasonable default if config is not available
maxLatency := 500 * time.Millisecond
if config := GetConfig(); config != nil {
maxLatency = config.MaxLatency
}
if latency > maxLatency {
return ErrInvalidLatency
}
return nil
}
// ValidateMetricsInterval validates metrics update interval
func ValidateMetricsInterval(interval time.Duration) error {
// Use reasonable defaults if config is not available
minInterval := 100 * time.Millisecond
maxInterval := 10 * time.Second
if config := GetConfig(); config != nil {
minInterval = config.MinMetricsUpdateInterval
maxInterval = config.MaxMetricsUpdateInterval
}
if interval < minInterval {
return ErrInvalidMetricsInterval
}
if interval > maxInterval {
return ErrInvalidMetricsInterval
}
return nil
}
// ValidateAdaptiveBufferConfig validates adaptive buffer configuration
func ValidateAdaptiveBufferConfig(minSize, maxSize, defaultSize int) error {
if minSize <= 0 || maxSize <= 0 || defaultSize <= 0 {
return ErrInvalidBufferSize
}
if minSize >= maxSize {
return ErrInvalidBufferSize
}
if defaultSize < minSize || defaultSize > maxSize {
return ErrInvalidBufferSize
}
// Validate against global limits
maxBuffer := 262144 // 256KB default
if config := GetConfig(); config != nil {
maxBuffer = config.SocketMaxBuffer
}
if maxSize > maxBuffer {
return ErrInvalidBufferSize
}
return nil
}
// ValidateInputIPCConfig validates input IPC configuration
func ValidateInputIPCConfig(sampleRate, channels, frameSize int) error {
// Use reasonable defaults if config is not available
minSampleRate := 8000
maxSampleRate := 48000
maxChannels := 8
if config := GetConfig(); config != nil {
minSampleRate = config.MinSampleRate
maxSampleRate = config.MaxSampleRate
maxChannels = config.MaxChannels
}
if sampleRate < minSampleRate || sampleRate > maxSampleRate {
return ErrInvalidSampleRate
}
if channels < 1 || channels > maxChannels {
return ErrInvalidChannels
}
if frameSize <= 0 {
return ErrInvalidFrameSize
}
return nil
}

View File

@ -0,0 +1,290 @@
package audio
import (
"errors"
"fmt"
"time"
"unsafe"
"github.com/rs/zerolog"
)
// Enhanced validation errors with more specific context
var (
ErrInvalidFrameLength = errors.New("invalid frame length")
ErrFrameDataCorrupted = errors.New("frame data appears corrupted")
ErrBufferAlignment = errors.New("buffer alignment invalid")
ErrInvalidSampleFormat = errors.New("invalid sample format")
ErrInvalidTimestamp = errors.New("invalid timestamp")
ErrConfigurationMismatch = errors.New("configuration mismatch")
ErrResourceExhaustion = errors.New("resource exhaustion detected")
ErrInvalidPointer = errors.New("invalid pointer")
ErrBufferOverflow = errors.New("buffer overflow detected")
ErrInvalidState = errors.New("invalid state")
)
// ValidationLevel defines the level of validation to perform
type ValidationLevel int
const (
ValidationMinimal ValidationLevel = iota // Only critical safety checks
ValidationStandard // Standard validation for production
ValidationStrict // Comprehensive validation for debugging
)
// ValidationConfig controls validation behavior
type ValidationConfig struct {
Level ValidationLevel
EnableRangeChecks bool
EnableAlignmentCheck bool
EnableDataIntegrity bool
MaxValidationTime time.Duration
}
// GetValidationConfig returns the current validation configuration
func GetValidationConfig() ValidationConfig {
return ValidationConfig{
Level: ValidationStandard,
EnableRangeChecks: true,
EnableAlignmentCheck: true,
EnableDataIntegrity: false, // Disabled by default for performance
MaxValidationTime: 5 * time.Second, // Default validation timeout
}
}
// ValidateAudioFrameFast performs minimal validation for performance-critical paths
func ValidateAudioFrameFast(data []byte) error {
if len(data) == 0 {
return ErrInvalidFrameData
}
// Quick bounds check using config constants
maxSize := GetConfig().MaxAudioFrameSize
if len(data) > maxSize {
return fmt.Errorf("%w: frame size %d exceeds maximum %d", ErrInvalidFrameSize, len(data), maxSize)
}
return nil
}
// ValidateAudioFrameComprehensive performs thorough validation
func ValidateAudioFrameComprehensive(data []byte, expectedSampleRate int, expectedChannels int) error {
validationConfig := GetValidationConfig()
start := time.Now()
// Timeout protection for validation
defer func() {
if time.Since(start) > validationConfig.MaxValidationTime {
// Log validation timeout but don't fail
getValidationLogger().Warn().Dur("duration", time.Since(start)).Msg("validation timeout exceeded")
}
}()
// Basic validation first
if err := ValidateAudioFrameFast(data); err != nil {
return err
}
// Range validation
if validationConfig.EnableRangeChecks {
config := GetConfig()
minFrameSize := 64 // Minimum reasonable frame size
if len(data) < minFrameSize {
return fmt.Errorf("%w: frame size %d below minimum %d", ErrInvalidFrameSize, len(data), minFrameSize)
}
// Validate frame length matches expected sample format
expectedFrameSize := (expectedSampleRate * expectedChannels * 2) / 1000 * int(config.AudioQualityMediumFrameSize/time.Millisecond)
tolerance := 512 // Frame size tolerance in bytes
if abs(len(data)-expectedFrameSize) > tolerance {
return fmt.Errorf("%w: frame size %d doesn't match expected %d (±%d)", ErrInvalidFrameLength, len(data), expectedFrameSize, tolerance)
}
}
// Alignment validation for ARM32 compatibility
if validationConfig.EnableAlignmentCheck {
if uintptr(unsafe.Pointer(&data[0]))%4 != 0 {
return fmt.Errorf("%w: buffer not 4-byte aligned for ARM32", ErrBufferAlignment)
}
}
// Data integrity checks (expensive, only for debugging)
if validationConfig.EnableDataIntegrity && validationConfig.Level == ValidationStrict {
if err := validateAudioDataIntegrity(data, expectedChannels); err != nil {
return err
}
}
return nil
}
// ValidateZeroCopyFrameEnhanced performs enhanced zero-copy frame validation
func ValidateZeroCopyFrameEnhanced(frame *ZeroCopyAudioFrame) error {
if frame == nil {
return fmt.Errorf("%w: frame is nil", ErrInvalidPointer)
}
// Check reference count validity
frame.mutex.RLock()
refCount := frame.refCount
length := frame.length
capacity := frame.capacity
frame.mutex.RUnlock()
if refCount <= 0 {
return fmt.Errorf("%w: invalid reference count %d", ErrInvalidState, refCount)
}
if length < 0 || capacity < 0 {
return fmt.Errorf("%w: negative length (%d) or capacity (%d)", ErrInvalidState, length, capacity)
}
if length > capacity {
return fmt.Errorf("%w: length %d exceeds capacity %d", ErrBufferOverflow, length, capacity)
}
// Validate the underlying data
data := frame.Data()
return ValidateAudioFrameFast(data)
}
// ValidateBufferBounds performs bounds checking with overflow protection
func ValidateBufferBounds(buffer []byte, offset, length int) error {
if buffer == nil {
return fmt.Errorf("%w: buffer is nil", ErrInvalidPointer)
}
if offset < 0 {
return fmt.Errorf("%w: negative offset %d", ErrInvalidState, offset)
}
if length < 0 {
return fmt.Errorf("%w: negative length %d", ErrInvalidState, length)
}
// Check for integer overflow
if offset > len(buffer) {
return fmt.Errorf("%w: offset %d exceeds buffer length %d", ErrBufferOverflow, offset, len(buffer))
}
// Safe addition check for overflow
if offset+length < offset || offset+length > len(buffer) {
return fmt.Errorf("%w: range [%d:%d] exceeds buffer length %d", ErrBufferOverflow, offset, offset+length, len(buffer))
}
return nil
}
// ValidateAudioConfiguration performs comprehensive configuration validation
func ValidateAudioConfiguration(config AudioConfig) error {
if err := ValidateAudioQuality(config.Quality); err != nil {
return fmt.Errorf("quality validation failed: %w", err)
}
configConstants := GetConfig()
// Validate bitrate ranges
minBitrate := 6000 // Minimum Opus bitrate
maxBitrate := 510000 // Maximum Opus bitrate
if config.Bitrate < minBitrate || config.Bitrate > maxBitrate {
return fmt.Errorf("%w: bitrate %d outside valid range [%d, %d]", ErrInvalidConfiguration, config.Bitrate, minBitrate, maxBitrate)
}
// Validate sample rate
validSampleRates := []int{8000, 12000, 16000, 24000, 48000}
validSampleRate := false
for _, rate := range validSampleRates {
if config.SampleRate == rate {
validSampleRate = true
break
}
}
if !validSampleRate {
return fmt.Errorf("%w: sample rate %d not in supported rates %v", ErrInvalidSampleRate, config.SampleRate, validSampleRates)
}
// Validate channels
if config.Channels < 1 || config.Channels > configConstants.MaxChannels {
return fmt.Errorf("%w: channels %d outside valid range [1, %d]", ErrInvalidChannels, config.Channels, configConstants.MaxChannels)
}
// Validate frame size
minFrameSize := 10 * time.Millisecond // Minimum frame duration
maxFrameSize := 100 * time.Millisecond // Maximum frame duration
if config.FrameSize < minFrameSize || config.FrameSize > maxFrameSize {
return fmt.Errorf("%w: frame size %v outside valid range [%v, %v]", ErrInvalidConfiguration, config.FrameSize, minFrameSize, maxFrameSize)
}
return nil
}
// ValidateResourceLimits checks if system resources are within acceptable limits
func ValidateResourceLimits() error {
config := GetConfig()
// Check buffer pool sizes
framePoolStats := GetAudioBufferPoolStats()
if framePoolStats.FramePoolSize > int64(config.MaxPoolSize*2) {
return fmt.Errorf("%w: frame pool size %d exceeds safe limit %d", ErrResourceExhaustion, framePoolStats.FramePoolSize, config.MaxPoolSize*2)
}
// Check zero-copy pool allocation count
zeroCopyStats := GetGlobalZeroCopyPoolStats()
if zeroCopyStats.AllocationCount > int64(config.MaxPoolSize*3) {
return fmt.Errorf("%w: zero-copy allocations %d exceed safe limit %d", ErrResourceExhaustion, zeroCopyStats.AllocationCount, config.MaxPoolSize*3)
}
return nil
}
// validateAudioDataIntegrity performs expensive data integrity checks
func validateAudioDataIntegrity(data []byte, channels int) error {
if len(data)%2 != 0 {
return fmt.Errorf("%w: odd number of bytes for 16-bit samples", ErrInvalidSampleFormat)
}
if len(data)%(channels*2) != 0 {
return fmt.Errorf("%w: data length %d not aligned to channel count %d", ErrInvalidSampleFormat, len(data), channels)
}
// Check for obvious corruption patterns (all zeros, all max values)
sampleCount := len(data) / 2
zeroCount := 0
maxCount := 0
for i := 0; i < len(data); i += 2 {
sample := int16(data[i]) | int16(data[i+1])<<8
switch sample {
case 0:
zeroCount++
case 32767, -32768:
maxCount++
}
}
// Flag suspicious patterns
if zeroCount > sampleCount*9/10 {
return fmt.Errorf("%w: %d%% zero samples suggests silence or corruption", ErrFrameDataCorrupted, (zeroCount*100)/sampleCount)
}
if maxCount > sampleCount/10 {
return fmt.Errorf("%w: %d%% max-value samples suggests clipping or corruption", ErrFrameDataCorrupted, (maxCount*100)/sampleCount)
}
return nil
}
// Helper function for absolute value
func abs(x int) int {
if x < 0 {
return -x
}
return x
}
// getValidationLogger returns a logger for validation operations
func getValidationLogger() *zerolog.Logger {
// Return a basic logger for validation
logger := zerolog.New(nil).With().Timestamp().Logger()
return &logger
}

View File

@ -7,8 +7,38 @@ import (
"unsafe"
)
// ZeroCopyAudioFrame represents an audio frame that can be passed between
// components without copying the underlying data
// ZeroCopyAudioFrame represents a reference-counted audio frame for zero-copy operations.
//
// This structure implements a sophisticated memory management system designed to minimize
// allocations and memory copying in the audio pipeline:
//
// Key Features:
//
// 1. Reference Counting: Multiple components can safely share the same frame data
// without copying. The frame is automatically returned to the pool when the last
// reference is released.
//
// 2. Thread Safety: All operations are protected by RWMutex, allowing concurrent
// reads while ensuring exclusive access for modifications.
//
// 3. Pool Integration: Frames are automatically managed by ZeroCopyFramePool,
// enabling efficient reuse and preventing memory fragmentation.
//
// 4. Unsafe Pointer Access: For performance-critical CGO operations, direct
// memory access is provided while maintaining safety through reference counting.
//
// Usage Pattern:
//
// frame := pool.Get() // Acquire frame (refCount = 1)
// frame.AddRef() // Share with another component (refCount = 2)
// data := frame.Data() // Access data safely
// frame.Release() // Release reference (refCount = 1)
// frame.Release() // Final release, returns to pool (refCount = 0)
//
// Memory Safety:
// - Frames cannot be modified while shared (refCount > 1)
// - Data access is bounds-checked to prevent buffer overruns
// - Pool management prevents use-after-free scenarios
type ZeroCopyAudioFrame struct {
data []byte
length int
@ -18,7 +48,37 @@ type ZeroCopyAudioFrame struct {
pooled bool
}
// ZeroCopyFramePool manages reusable zero-copy audio frames
// ZeroCopyFramePool manages a pool of reusable zero-copy audio frames.
//
// This pool implements a three-tier memory management strategy optimized for
// real-time audio processing with minimal allocation overhead:
//
// Tier 1 - Pre-allocated Frames:
//
// A small number of frames are pre-allocated at startup and kept ready
// for immediate use. This provides the fastest possible allocation for
// the most common case and eliminates allocation latency spikes.
//
// Tier 2 - sync.Pool Cache:
//
// The standard Go sync.Pool provides efficient reuse of frames with
// automatic garbage collection integration. Frames are automatically
// returned here when memory pressure is low.
//
// Tier 3 - Memory Guard:
//
// A configurable limit prevents excessive memory usage by limiting
// the total number of allocated frames. When the limit is reached,
// allocation requests are denied to prevent OOM conditions.
//
// Performance Characteristics:
// - Pre-allocated tier: ~10ns allocation time
// - sync.Pool tier: ~50ns allocation time
// - Memory guard: Prevents unbounded growth
// - Metrics tracking: Hit/miss rates for optimization
//
// The pool is designed for embedded systems with limited memory (256MB)
// where predictable memory usage is more important than absolute performance.
type ZeroCopyFramePool struct {
// Atomic fields MUST be first for ARM32 alignment (int64 fields need 8-byte alignment)
counter int64 // Frame counter (atomic)

View File

@ -967,9 +967,7 @@ func rpcSetUsbDevices(usbDevices usbgadget.Devices) error {
// Stop audio output supervisor
if audioSupervisor != nil && audioSupervisor.IsRunning() {
logger.Info().Msg("stopping audio output supervisor")
if err := audioSupervisor.Stop(); err != nil {
logger.Error().Err(err).Msg("failed to stop audio supervisor")
}
audioSupervisor.Stop()
// Wait for audio processes to fully stop before proceeding
for i := 0; i < 50; i++ { // Wait up to 5 seconds
if !audioSupervisor.IsRunning() {
@ -1063,9 +1061,7 @@ func rpcSetUsbDeviceState(device string, enabled bool) error {
// Stop audio output supervisor
if audioSupervisor != nil && audioSupervisor.IsRunning() {
logger.Info().Msg("stopping audio output supervisor")
if err := audioSupervisor.Stop(); err != nil {
logger.Error().Err(err).Msg("failed to stop audio supervisor")
}
audioSupervisor.Stop()
// Wait for audio processes to fully stop
for i := 0; i < 50; i++ { // Wait up to 5 seconds
if !audioSupervisor.IsRunning() {

View File

@ -18,7 +18,7 @@ var (
appCtx context.Context
isAudioServer bool
audioProcessDone chan struct{}
audioSupervisor *audio.AudioServerSupervisor
audioSupervisor *audio.AudioOutputSupervisor
)
// runAudioServer is now handled by audio.RunAudioOutputServer
@ -36,7 +36,7 @@ func startAudioSubprocess() error {
audio.StartAdaptiveBuffering()
// Create audio server supervisor
audioSupervisor = audio.NewAudioServerSupervisor()
audioSupervisor = audio.NewAudioOutputSupervisor()
// Set the global supervisor for access from audio package
audio.SetAudioOutputSupervisor(audioSupervisor)
@ -251,9 +251,7 @@ func Main(audioServer bool, audioInputServer bool) {
if !isAudioServer {
if audioSupervisor != nil {
logger.Info().Msg("stopping audio supervisor")
if err := audioSupervisor.Stop(); err != nil {
logger.Error().Err(err).Msg("failed to stop audio supervisor")
}
audioSupervisor.Stop()
}
<-audioProcessDone
} else {

View File

@ -9,6 +9,8 @@ import { useMicrophone } from "@/hooks/useMicrophone";
import { useAudioLevel } from "@/hooks/useAudioLevel";
import { useAudioEvents } from "@/hooks/useAudioEvents";
import api from "@/api";
import { AUDIO_CONFIG } from "@/config/constants";
import audioQualityService from "@/services/audioQualityService";
interface AudioMetrics {
frames_received: number;
@ -44,12 +46,8 @@ interface AudioConfig {
FrameSize: string;
}
const qualityLabels = {
0: "Low",
1: "Medium",
2: "High",
3: "Ultra"
};
// Quality labels will be managed by the audio quality service
const getQualityLabels = () => audioQualityService.getQualityLabels();
// Format percentage values to 2 decimal places
function formatPercentage(value: number | null | undefined): string {
@ -246,22 +244,15 @@ export default function AudioMetricsDashboard() {
const loadAudioConfig = async () => {
try {
// Load config
const configResp = await api.GET("/audio/quality");
if (configResp.ok) {
const configData = await configResp.json();
setConfig(configData.current);
// Use centralized audio quality service
const { audio, microphone } = await audioQualityService.loadAllConfigurations();
if (audio) {
setConfig(audio.current);
}
// Load microphone config
try {
const micConfigResp = await api.GET("/microphone/quality");
if (micConfigResp.ok) {
const micConfigData = await micConfigResp.json();
setMicrophoneConfig(micConfigData.current);
}
} catch {
// Microphone config not available
if (microphone) {
setMicrophoneConfig(microphone.current);
}
} catch (error) {
console.error("Failed to load audio config:", error);
@ -397,7 +388,7 @@ export default function AudioMetricsDashboard() {
const getDropRate = () => {
if (!metrics || metrics.frames_received === 0) return 0;
return ((metrics.frames_dropped / metrics.frames_received) * 100);
return ((metrics.frames_dropped / metrics.frames_received) * AUDIO_CONFIG.PERCENTAGE_MULTIPLIER);
};
@ -449,7 +440,7 @@ export default function AudioMetricsDashboard() {
<div className="flex justify-between">
<span className="text-slate-500 dark:text-slate-400">Quality:</span>
<span className={cx("font-medium", getQualityColor(config.Quality))}>
{qualityLabels[config.Quality as keyof typeof qualityLabels]}
{getQualityLabels()[config.Quality]}
</span>
</div>
<div className="flex justify-between">
@ -486,7 +477,7 @@ export default function AudioMetricsDashboard() {
<div className="flex justify-between">
<span className="text-slate-500 dark:text-slate-400">Quality:</span>
<span className={cx("font-medium", getQualityColor(microphoneConfig.Quality))}>
{qualityLabels[microphoneConfig.Quality as keyof typeof qualityLabels]}
{getQualityLabels()[microphoneConfig.Quality]}
</span>
</div>
<div className="flex justify-between">
@ -668,26 +659,26 @@ export default function AudioMetricsDashboard() {
</span>
<span className={cx(
"font-bold",
getDropRate() > 5
getDropRate() > AUDIO_CONFIG.DROP_RATE_CRITICAL_THRESHOLD
? "text-red-600 dark:text-red-400"
: getDropRate() > 1
: getDropRate() > AUDIO_CONFIG.DROP_RATE_WARNING_THRESHOLD
? "text-yellow-600 dark:text-yellow-400"
: "text-green-600 dark:text-green-400"
)}>
{getDropRate().toFixed(2)}%
{getDropRate().toFixed(AUDIO_CONFIG.PERCENTAGE_DECIMAL_PLACES)}%
</span>
</div>
<div className="mt-1 h-2 w-full rounded-full bg-slate-200 dark:bg-slate-600">
<div
className={cx(
"h-2 rounded-full transition-all duration-300",
getDropRate() > 5
getDropRate() > AUDIO_CONFIG.DROP_RATE_CRITICAL_THRESHOLD
? "bg-red-500"
: getDropRate() > 1
: getDropRate() > AUDIO_CONFIG.DROP_RATE_WARNING_THRESHOLD
? "bg-yellow-500"
: "bg-green-500"
)}
style={{ width: `${Math.min(getDropRate(), 100)}%` }}
style={{ width: `${Math.min(getDropRate(), AUDIO_CONFIG.MAX_LEVEL_PERCENTAGE)}%` }}
/>
</div>
</div>
@ -734,27 +725,27 @@ export default function AudioMetricsDashboard() {
</span>
<span className={cx(
"font-bold",
(microphoneMetrics.frames_sent > 0 ? (microphoneMetrics.frames_dropped / microphoneMetrics.frames_sent) * 100 : 0) > 5
(microphoneMetrics.frames_sent > 0 ? (microphoneMetrics.frames_dropped / microphoneMetrics.frames_sent) * AUDIO_CONFIG.PERCENTAGE_MULTIPLIER : 0) > AUDIO_CONFIG.DROP_RATE_CRITICAL_THRESHOLD
? "text-red-600 dark:text-red-400"
: (microphoneMetrics.frames_sent > 0 ? (microphoneMetrics.frames_dropped / microphoneMetrics.frames_sent) * 100 : 0) > 1
: (microphoneMetrics.frames_sent > 0 ? (microphoneMetrics.frames_dropped / microphoneMetrics.frames_sent) * AUDIO_CONFIG.PERCENTAGE_MULTIPLIER : 0) > AUDIO_CONFIG.DROP_RATE_WARNING_THRESHOLD
? "text-yellow-600 dark:text-yellow-400"
: "text-green-600 dark:text-green-400"
)}>
{microphoneMetrics.frames_sent > 0 ? ((microphoneMetrics.frames_dropped / microphoneMetrics.frames_sent) * 100).toFixed(2) : "0.00"}%
{microphoneMetrics.frames_sent > 0 ? ((microphoneMetrics.frames_dropped / microphoneMetrics.frames_sent) * AUDIO_CONFIG.PERCENTAGE_MULTIPLIER).toFixed(AUDIO_CONFIG.PERCENTAGE_DECIMAL_PLACES) : "0.00"}%
</span>
</div>
<div className="mt-1 h-2 w-full rounded-full bg-slate-200 dark:bg-slate-600">
<div
className={cx(
"h-2 rounded-full transition-all duration-300",
(microphoneMetrics.frames_sent > 0 ? (microphoneMetrics.frames_dropped / microphoneMetrics.frames_sent) * 100 : 0) > 5
(microphoneMetrics.frames_sent > 0 ? (microphoneMetrics.frames_dropped / microphoneMetrics.frames_sent) * AUDIO_CONFIG.PERCENTAGE_MULTIPLIER : 0) > AUDIO_CONFIG.DROP_RATE_CRITICAL_THRESHOLD
? "bg-red-500"
: (microphoneMetrics.frames_sent > 0 ? (microphoneMetrics.frames_dropped / microphoneMetrics.frames_sent) * 100 : 0) > 1
: (microphoneMetrics.frames_sent > 0 ? (microphoneMetrics.frames_dropped / microphoneMetrics.frames_sent) * AUDIO_CONFIG.PERCENTAGE_MULTIPLIER : 0) > AUDIO_CONFIG.DROP_RATE_WARNING_THRESHOLD
? "bg-yellow-500"
: "bg-green-500"
)}
style={{
width: `${Math.min(microphoneMetrics.frames_sent > 0 ? (microphoneMetrics.frames_dropped / microphoneMetrics.frames_sent) * 100 : 0, 100)}%`
width: `${Math.min(microphoneMetrics.frames_sent > 0 ? (microphoneMetrics.frames_dropped / microphoneMetrics.frames_sent) * AUDIO_CONFIG.PERCENTAGE_MULTIPLIER : 0, AUDIO_CONFIG.MAX_LEVEL_PERCENTAGE)}%`
}}
/>
</div>

View File

@ -11,6 +11,8 @@ import { useAudioLevel } from "@/hooks/useAudioLevel";
import { useAudioEvents } from "@/hooks/useAudioEvents";
import api from "@/api";
import notifications from "@/notifications";
import { AUDIO_CONFIG } from "@/config/constants";
import audioQualityService from "@/services/audioQualityService";
// Type for microphone error
interface MicrophoneError {
@ -41,12 +43,8 @@ interface AudioConfig {
FrameSize: string;
}
const qualityLabels = {
0: "Low (32kbps)",
1: "Medium (64kbps)",
2: "High (128kbps)",
3: "Ultra (256kbps)"
};
// Quality labels will be managed by the audio quality service
const getQualityLabels = () => audioQualityService.getQualityLabels();
interface AudioControlPopoverProps {
microphone: MicrophoneHookReturn;
@ -138,20 +136,15 @@ export default function AudioControlPopover({ microphone, open }: AudioControlPo
const loadAudioConfigurations = async () => {
try {
// Parallel loading for better performance
const [qualityResp, micQualityResp] = await Promise.all([
api.GET("/audio/quality"),
api.GET("/microphone/quality")
]);
// Use centralized audio quality service
const { audio, microphone } = await audioQualityService.loadAllConfigurations();
if (qualityResp.ok) {
const qualityData = await qualityResp.json();
setCurrentConfig(qualityData.current);
if (audio) {
setCurrentConfig(audio.current);
}
if (micQualityResp.ok) {
const micQualityData = await micQualityResp.json();
setCurrentMicrophoneConfig(micQualityData.current);
if (microphone) {
setCurrentMicrophoneConfig(microphone.current);
}
setConfigsLoaded(true);
@ -511,7 +504,7 @@ export default function AudioControlPopover({ microphone, open }: AudioControlPo
</div>
<div className="grid grid-cols-2 gap-2">
{Object.entries(qualityLabels).map(([quality, label]) => (
{Object.entries(getQualityLabels()).map(([quality, label]) => (
<button
key={`mic-${quality}`}
onClick={() => handleMicrophoneQualityChange(parseInt(quality))}
@ -552,7 +545,7 @@ export default function AudioControlPopover({ microphone, open }: AudioControlPo
</div>
<div className="grid grid-cols-2 gap-2">
{Object.entries(qualityLabels).map(([quality, label]) => (
{Object.entries(getQualityLabels()).map(([quality, label]) => (
<button
key={quality}
onClick={() => handleQualityChange(parseInt(quality))}
@ -704,13 +697,13 @@ export default function AudioControlPopover({ microphone, open }: AudioControlPo
<div className="text-xs text-slate-500 dark:text-slate-400">Drop Rate</div>
<div className={cx(
"font-mono text-sm",
((metrics.frames_dropped / metrics.frames_received) * 100) > 5
((metrics.frames_dropped / metrics.frames_received) * AUDIO_CONFIG.PERCENTAGE_MULTIPLIER) > AUDIO_CONFIG.DROP_RATE_CRITICAL_THRESHOLD
? "text-red-600 dark:text-red-400"
: ((metrics.frames_dropped / metrics.frames_received) * 100) > 1
: ((metrics.frames_dropped / metrics.frames_received) * AUDIO_CONFIG.PERCENTAGE_MULTIPLIER) > AUDIO_CONFIG.DROP_RATE_WARNING_THRESHOLD
? "text-yellow-600 dark:text-yellow-400"
: "text-green-600 dark:text-green-400"
)}>
{((metrics.frames_dropped / metrics.frames_received) * 100).toFixed(2)}%
{((metrics.frames_dropped / metrics.frames_received) * AUDIO_CONFIG.PERCENTAGE_MULTIPLIER).toFixed(AUDIO_CONFIG.PERCENTAGE_DECIMAL_PLACES)}%
</div>
</div>
)}

167
ui/src/config/constants.ts Normal file
View File

@ -0,0 +1,167 @@
// Centralized configuration constants
// Network and API Configuration
export const NETWORK_CONFIG = {
WEBSOCKET_RECONNECT_INTERVAL: 3000,
LONG_PRESS_DURATION: 3000,
ERROR_MESSAGE_TIMEOUT: 3000,
AUDIO_TEST_DURATION: 5000,
BACKEND_RETRY_DELAY: 500,
RESET_DELAY: 200,
STATE_CHECK_DELAY: 100,
VERIFICATION_DELAY: 1000,
} as const;
// Default URLs and Endpoints
export const DEFAULT_URLS = {
JETKVM_PROD_API: "https://api.jetkvm.com",
JETKVM_PROD_APP: "https://app.jetkvm.com",
JETKVM_DOCS_TROUBLESHOOTING: "https://jetkvm.com/docs/getting-started/troubleshooting",
JETKVM_DOCS_REMOTE_ACCESS: "https://jetkvm.com/docs/networking/remote-access",
JETKVM_DOCS_LOCAL_ACCESS_RESET: "https://jetkvm.com/docs/networking/local-access#reset-password",
JETKVM_GITHUB: "https://github.com/jetkvm",
CRONTAB_GURU: "https://crontab.guru/examples.html",
} as const;
// Sample ISO URLs for mounting
export const SAMPLE_ISOS = {
UBUNTU_24_04: {
name: "Ubuntu 24.04.2 Desktop",
url: "https://releases.ubuntu.com/24.04.2/ubuntu-24.04.2-desktop-amd64.iso",
},
DEBIAN_13: {
name: "Debian 13.0.0 (Testing)",
url: "https://cdimage.debian.org/debian-cd/current/amd64/iso-cd/debian-13.0.0-amd64-netinst.iso",
},
DEBIAN_12: {
name: "Debian 12.11.0 (Stable)",
url: "https://cdimage.debian.org/mirror/cdimage/archive/12.11.0/amd64/iso-cd/debian-12.11.0-amd64-netinst.iso",
},
FEDORA_41: {
name: "Fedora 41 Workstation",
url: "https://download.fedoraproject.org/pub/fedora/linux/releases/41/Workstation/x86_64/iso/Fedora-Workstation-Live-x86_64-41-1.4.iso",
},
OPENSUSE_LEAP: {
name: "openSUSE Leap 15.6",
url: "https://download.opensuse.org/distribution/leap/15.6/iso/openSUSE-Leap-15.6-NET-x86_64-Media.iso",
},
OPENSUSE_TUMBLEWEED: {
name: "openSUSE Tumbleweed",
url: "https://download.opensuse.org/tumbleweed/iso/openSUSE-Tumbleweed-NET-x86_64-Current.iso",
},
ARCH_LINUX: {
name: "Arch Linux",
url: "https://archlinux.doridian.net/iso/2025.02.01/archlinux-2025.02.01-x86_64.iso",
},
NETBOOT_XYZ: {
name: "netboot.xyz",
url: "https://boot.netboot.xyz/ipxe/netboot.xyz.iso",
},
} as const;
// Security and Access Configuration
export const SECURITY_CONFIG = {
LOCALHOST_ONLY_IP: "127.0.0.1",
LOCALHOST_HOSTNAME: "localhost",
HTTPS_PROTOCOL: "https:",
} as const;
// Default Hardware Configuration
export const HARDWARE_CONFIG = {
DEFAULT_OFF_AFTER: 50000,
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
} as const;
// Audio Configuration
export const AUDIO_CONFIG = {
// Audio Level Analysis
LEVEL_UPDATE_INTERVAL: 100, // ms - throttle audio level updates for performance
FFT_SIZE: 128, // reduced from 256 for better performance
SMOOTHING_TIME_CONSTANT: 0.8,
RELEVANT_FREQUENCY_BINS: 32, // focus on lower frequencies for voice
RMS_SCALING_FACTOR: 180, // for converting RMS to percentage
MAX_LEVEL_PERCENTAGE: 100,
// Microphone Configuration
SAMPLE_RATE: 48000, // Hz - high quality audio sampling
CHANNEL_COUNT: 1, // mono for microphone input
OPERATION_DEBOUNCE_MS: 1000, // debounce microphone operations
SYNC_DEBOUNCE_MS: 1000, // debounce state synchronization
AUDIO_TEST_TIMEOUT: 100, // ms - timeout for audio testing
// Audio Output Quality Bitrates (matching backend config_constants.go)
OUTPUT_QUALITY_BITRATES: {
LOW: 32, // AudioQualityLowOutputBitrate
MEDIUM: 64, // AudioQualityMediumOutputBitrate
HIGH: 128, // AudioQualityHighOutputBitrate
ULTRA: 192, // AudioQualityUltraOutputBitrate
} as const,
// Audio Input Quality Bitrates (matching backend config_constants.go)
INPUT_QUALITY_BITRATES: {
LOW: 16, // AudioQualityLowInputBitrate
MEDIUM: 32, // AudioQualityMediumInputBitrate
HIGH: 64, // AudioQualityHighInputBitrate
ULTRA: 96, // AudioQualityUltraInputBitrate
} as const,
// Sample Rates (matching backend config_constants.go)
QUALITY_SAMPLE_RATES: {
LOW: 22050, // AudioQualityLowSampleRate
MEDIUM: 44100, // AudioQualityMediumSampleRate
HIGH: 48000, // Default SampleRate
ULTRA: 48000, // Default SampleRate
} as const,
// Microphone Sample Rates
MIC_QUALITY_SAMPLE_RATES: {
LOW: 16000, // AudioQualityMicLowSampleRate
MEDIUM: 44100, // AudioQualityMediumSampleRate
HIGH: 48000, // Default SampleRate
ULTRA: 48000, // Default SampleRate
} as const,
// Channels (matching backend config_constants.go)
QUALITY_CHANNELS: {
LOW: 1, // AudioQualityLowChannels (mono)
MEDIUM: 2, // AudioQualityMediumChannels (stereo)
HIGH: 2, // AudioQualityHighChannels (stereo)
ULTRA: 2, // AudioQualityUltraChannels (stereo)
} as const,
// Frame Sizes in milliseconds (matching backend config_constants.go)
QUALITY_FRAME_SIZES: {
LOW: 40, // AudioQualityLowFrameSize (40ms)
MEDIUM: 20, // AudioQualityMediumFrameSize (20ms)
HIGH: 20, // AudioQualityHighFrameSize (20ms)
ULTRA: 10, // AudioQualityUltraFrameSize (10ms)
} as const,
// Updated Quality Labels with correct output bitrates
QUALITY_LABELS: {
0: "Low (32 kbps)",
1: "Medium (64 kbps)",
2: "High (128 kbps)",
3: "Ultra (192 kbps)",
} as const,
// Legacy support - keeping for backward compatibility
QUALITY_BITRATES: {
LOW: 32,
MEDIUM: 64,
HIGH: 128,
ULTRA: 192, // Updated to match backend
},
// Audio Analysis
ANALYSIS_FFT_SIZE: 256, // for detailed audio analysis
ANALYSIS_UPDATE_INTERVAL: 100, // ms - 10fps for audio level updates
LEVEL_SCALING_FACTOR: 255, // for RMS to percentage conversion
// Audio Metrics Thresholds
DROP_RATE_WARNING_THRESHOLD: 1, // percentage - yellow warning
DROP_RATE_CRITICAL_THRESHOLD: 5, // percentage - red critical
PERCENTAGE_MULTIPLIER: 100, // for converting ratios to percentages
PERCENTAGE_DECIMAL_PLACES: 2, // decimal places for percentage display
} as const;
// Placeholder URLs
export const PLACEHOLDERS = {
ISO_URL: "https://example.com/image.iso",
PROXY_URL: "http://proxy.example.com:8080/",
API_URL: "https://api.example.com",
APP_URL: "https://app.example.com",
} as const;

View File

@ -7,6 +7,8 @@ import {
MAX_KEYS_PER_STEP,
} from "@/constants/macros";
import { devWarn } from '../utils/debug';
// Define the JsonRpc types for better type checking
interface JsonRpcResponse {
jsonrpc: string;
@ -782,7 +784,7 @@ export const useNetworkStateStore = create<NetworkState>((set, get) => ({
setDhcpLeaseExpiry: (expiry: Date) => {
const lease = get().dhcp_lease;
if (!lease) {
console.warn("No lease found");
devWarn("No lease found");
return;
}

View File

@ -2,6 +2,7 @@ import { useNavigate, useParams, NavigateOptions } from "react-router-dom";
import { useCallback, useMemo } from "react";
import { isOnDevice } from "../main";
import { devError } from '../utils/debug';
/**
* Generates the correct path based on whether the app is running on device or in cloud mode
@ -21,7 +22,7 @@ export function getDeviceUiPath(path: string, deviceId?: string): string {
return normalizedPath;
} else {
if (!deviceId) {
console.error("No device ID provided when generating path in cloud mode");
devError("No device ID provided when generating path in cloud mode");
throw new Error("Device ID is required for cloud mode path generation");
}
return `/devices/${deviceId}${normalizedPath}`;

View File

@ -1,5 +1,7 @@
import { useState, useEffect, useCallback } from 'react';
import { devError } from '../utils/debug';
export interface AudioDevice {
deviceId: string;
label: string;
@ -66,7 +68,7 @@ export function useAudioDevices(): UseAudioDevicesReturn {
// Audio devices enumerated
} catch (err) {
console.error('Failed to enumerate audio devices:', err);
devError('Failed to enumerate audio devices:', err);
setError(err instanceof Error ? err.message : 'Failed to access audio devices');
} finally {
setIsLoading(false);

View File

@ -1,6 +1,9 @@
import { useCallback, useEffect, useRef, useState } from 'react';
import useWebSocket, { ReadyState } from 'react-use-websocket';
import { devError, devWarn } from '../utils/debug';
import { NETWORK_CONFIG } from '../config/constants';
// Audio event types matching the backend
export type AudioEventType =
| 'audio-mute-changed'
@ -121,7 +124,7 @@ export function useAudioEvents(onAudioDeviceChanged?: (data: AudioDeviceChangedD
} = useWebSocket(getWebSocketUrl(), {
shouldReconnect: () => true,
reconnectAttempts: 10,
reconnectInterval: 3000,
reconnectInterval: NETWORK_CONFIG.WEBSOCKET_RECONNECT_INTERVAL,
share: true, // Share the WebSocket connection across multiple hooks
onOpen: () => {
// WebSocket connected
@ -137,7 +140,7 @@ export function useAudioEvents(onAudioDeviceChanged?: (data: AudioDeviceChangedD
globalSubscriptionState.connectionId = null;
},
onError: (event) => {
console.error('[AudioEvents] WebSocket error:', event);
devError('[AudioEvents] WebSocket error:', event);
},
});
@ -270,7 +273,7 @@ export function useAudioEvents(onAudioDeviceChanged?: (data: AudioDeviceChangedD
} catch (error) {
// Ignore parsing errors for non-JSON messages (like "pong")
if (lastMessage.data !== 'pong') {
console.warn('[AudioEvents] Failed to parse WebSocket message:', error);
devWarn('[AudioEvents] Failed to parse WebSocket message:', error);
}
}
}

View File

@ -1,5 +1,7 @@
import { useEffect, useRef, useState } from 'react';
import { AUDIO_CONFIG } from '@/config/constants';
interface AudioLevelHookResult {
audioLevel: number; // 0-100 percentage
isAnalyzing: boolean;
@ -7,14 +9,14 @@ interface AudioLevelHookResult {
interface AudioLevelOptions {
enabled?: boolean; // Allow external control of analysis
updateInterval?: number; // Throttle updates (default: 100ms for 10fps instead of 60fps)
updateInterval?: number; // Throttle updates (default from AUDIO_CONFIG)
}
export const useAudioLevel = (
stream: MediaStream | null,
options: AudioLevelOptions = {}
): AudioLevelHookResult => {
const { enabled = true, updateInterval = 100 } = options;
const { enabled = true, updateInterval = AUDIO_CONFIG.LEVEL_UPDATE_INTERVAL } = options;
const [audioLevel, setAudioLevel] = useState(0);
const [isAnalyzing, setIsAnalyzing] = useState(false);
@ -59,8 +61,8 @@ export const useAudioLevel = (
const source = audioContext.createMediaStreamSource(stream);
// Configure analyser - use smaller FFT for better performance
analyser.fftSize = 128; // Reduced from 256 for better performance
analyser.smoothingTimeConstant = 0.8;
analyser.fftSize = AUDIO_CONFIG.FFT_SIZE;
analyser.smoothingTimeConstant = AUDIO_CONFIG.SMOOTHING_TIME_CONSTANT;
// Connect nodes
source.connect(analyser);
@ -87,7 +89,7 @@ export const useAudioLevel = (
// Optimized RMS calculation - process only relevant frequency bands
let sum = 0;
const relevantBins = Math.min(dataArray.length, 32); // Focus on lower frequencies for voice
const relevantBins = Math.min(dataArray.length, AUDIO_CONFIG.RELEVANT_FREQUENCY_BINS);
for (let i = 0; i < relevantBins; i++) {
const value = dataArray[i];
sum += value * value;
@ -95,7 +97,7 @@ export const useAudioLevel = (
const rms = Math.sqrt(sum / relevantBins);
// Convert to percentage (0-100) with better scaling
const level = Math.min(100, Math.max(0, (rms / 180) * 100)); // Adjusted scaling for better sensitivity
const level = Math.min(AUDIO_CONFIG.MAX_LEVEL_PERCENTAGE, Math.max(0, (rms / AUDIO_CONFIG.RMS_SCALING_FACTOR) * AUDIO_CONFIG.MAX_LEVEL_PERCENTAGE));
setAudioLevel(Math.round(level));
};

View File

@ -2,6 +2,8 @@ import { useCallback, useEffect } from "react";
import { useRTCStore } from "@/hooks/stores";
import { devError } from '../utils/debug';
export interface JsonRpcRequest {
jsonrpc: string;
method: string;
@ -61,7 +63,7 @@ export function useJsonRpc(onRequest?: (payload: JsonRpcRequest) => void) {
return;
}
if ("error" in payload) console.error(payload.error);
if ("error" in payload) devError(payload.error);
if (!payload.id) return;
const callback = callbackStore.get(payload.id);

View File

@ -2,6 +2,8 @@ import { useCallback, useEffect, useRef, useState } from "react";
import { useRTCStore } from "@/hooks/stores";
import api from "@/api";
import { devLog, devInfo, devWarn, devError, devOnly } from "@/utils/debug";
import { NETWORK_CONFIG, AUDIO_CONFIG } from "@/config/constants";
export interface MicrophoneError {
type: 'permission' | 'device' | 'network' | 'unknown';
@ -31,15 +33,14 @@ export function useMicrophone() {
// Add debouncing refs to prevent rapid operations
const lastOperationRef = useRef<number>(0);
const operationTimeoutRef = useRef<number | null>(null);
const OPERATION_DEBOUNCE_MS = 1000; // 1 second debounce
// Debounced operation wrapper
const debouncedOperation = useCallback((operation: () => Promise<void>, operationType: string) => {
const now = Date.now();
const timeSinceLastOp = now - lastOperationRef.current;
if (timeSinceLastOp < OPERATION_DEBOUNCE_MS) {
console.log(`Debouncing ${operationType} operation - too soon (${timeSinceLastOp}ms since last)`);
if (timeSinceLastOp < AUDIO_CONFIG.OPERATION_DEBOUNCE_MS) {
devLog(`Debouncing ${operationType} operation - too soon (${timeSinceLastOp}ms since last)`);
return;
}
@ -51,7 +52,7 @@ export function useMicrophone() {
lastOperationRef.current = now;
operation().catch(error => {
console.error(`Debounced ${operationType} operation failed:`, error);
devError(`Debounced ${operationType} operation failed:`, error);
});
}, []);
@ -72,7 +73,7 @@ export function useMicrophone() {
try {
await microphoneSender.replaceTrack(null);
} catch (error) {
console.warn("Failed to replace track with null:", error);
devWarn("Failed to replace track with null:", error);
// Fallback to removing the track
peerConnection.removeTrack(microphoneSender);
}
@ -110,14 +111,14 @@ export function useMicrophone() {
} : "No peer connection",
streamMatch: refStream === microphoneStream
};
console.log("Microphone Debug State:", state);
devLog("Microphone Debug State:", state);
// Also check if streams are active
if (refStream) {
console.log("Ref stream active tracks:", refStream.getAudioTracks().filter(t => t.readyState === 'live').length);
devLog("Ref stream active tracks:", refStream.getAudioTracks().filter(t => t.readyState === 'live').length);
}
if (microphoneStream && microphoneStream !== refStream) {
console.log("Store stream active tracks:", microphoneStream.getAudioTracks().filter(t => t.readyState === 'live').length);
devLog("Store stream active tracks:", microphoneStream.getAudioTracks().filter(t => t.readyState === 'live').length);
}
return state;
@ -137,15 +138,15 @@ export function useMicrophone() {
const syncMicrophoneState = useCallback(async () => {
// Debounce sync calls to prevent race conditions
const now = Date.now();
if (now - lastSyncRef.current < 1000) { // Increased debounce time
console.log("Skipping sync - too frequent");
if (now - lastSyncRef.current < AUDIO_CONFIG.SYNC_DEBOUNCE_MS) {
devLog("Skipping sync - too frequent");
return;
}
lastSyncRef.current = now;
// Don't sync if we're in the middle of starting the microphone
if (isStartingRef.current) {
console.log("Skipping sync - microphone is starting");
devLog("Skipping sync - microphone is starting");
return;
}
@ -157,27 +158,27 @@ export function useMicrophone() {
// Only sync if there's a significant state difference and we're not in a transition
if (backendRunning !== isMicrophoneActive) {
console.info(`Syncing microphone state: backend=${backendRunning}, frontend=${isMicrophoneActive}`);
devInfo(`Syncing microphone state: backend=${backendRunning}, frontend=${isMicrophoneActive}`);
// If backend is running but frontend thinks it's not, just update frontend state
if (backendRunning && !isMicrophoneActive) {
console.log("Backend running, updating frontend state to active");
devLog("Backend running, updating frontend state to active");
setMicrophoneActive(true);
}
// If backend is not running but frontend thinks it is, clean up and update state
else if (!backendRunning && isMicrophoneActive) {
console.log("Backend not running, cleaning up frontend state");
devLog("Backend not running, cleaning up frontend state");
setMicrophoneActive(false);
// Only clean up stream if we actually have one
if (microphoneStreamRef.current) {
console.log("Cleaning up orphaned stream");
devLog("Cleaning up orphaned stream");
await stopMicrophoneStream();
}
}
}
}
} catch (error) {
console.warn("Failed to sync microphone state:", error);
devWarn("Failed to sync microphone state:", error);
}
}, [isMicrophoneActive, setMicrophoneActive, stopMicrophoneStream]);
@ -185,7 +186,7 @@ export function useMicrophone() {
const startMicrophone = useCallback(async (deviceId?: string): Promise<{ success: boolean; error?: MicrophoneError }> => {
// Prevent multiple simultaneous start operations
if (isStarting || isStopping || isToggling) {
console.log("Microphone operation already in progress, skipping start");
devLog("Microphone operation already in progress, skipping start");
return { success: false, error: { type: 'unknown', message: 'Operation already in progress' } };
}
@ -198,8 +199,8 @@ export function useMicrophone() {
echoCancellation: true,
noiseSuppression: true,
autoGainControl: true,
sampleRate: 48000,
channelCount: 1,
sampleRate: AUDIO_CONFIG.SAMPLE_RATE,
channelCount: AUDIO_CONFIG.CHANNEL_COUNT,
};
// Add device ID if specified
@ -207,7 +208,7 @@ export function useMicrophone() {
audioConstraints.deviceId = { exact: deviceId };
}
console.log("Requesting microphone with constraints:", audioConstraints);
devLog("Requesting microphone with constraints:", audioConstraints);
const stream = await navigator.mediaDevices.getUserMedia({
audio: audioConstraints
});
@ -219,14 +220,14 @@ export function useMicrophone() {
setMicrophoneStream(stream);
// Verify the stream was stored correctly
console.log("Stream storage verification:", {
devLog("Stream storage verification:", {
refSet: !!microphoneStreamRef.current,
refId: microphoneStreamRef.current?.id,
storeWillBeSet: true // Store update is async
});
// Add audio track to peer connection if available
console.log("Peer connection state:", peerConnection ? {
devLog("Peer connection state:", peerConnection ? {
connectionState: peerConnection.connectionState,
iceConnectionState: peerConnection.iceConnectionState,
signalingState: peerConnection.signalingState
@ -234,11 +235,11 @@ export function useMicrophone() {
if (peerConnection && stream.getAudioTracks().length > 0) {
const audioTrack = stream.getAudioTracks()[0];
console.log("Starting microphone with audio track:", audioTrack.id, "kind:", audioTrack.kind);
devLog("Starting microphone with audio track:", audioTrack.id, "kind:", audioTrack.kind);
// Find the audio transceiver (should already exist with sendrecv direction)
const transceivers = peerConnection.getTransceivers();
console.log("Available transceivers:", transceivers.map(t => ({
devLog("Available transceivers:", transceivers.map(t => ({
direction: t.direction,
mid: t.mid,
senderTrack: t.sender.track?.kind,
@ -264,7 +265,7 @@ export function useMicrophone() {
return false;
});
console.log("Found audio transceiver:", audioTransceiver ? {
devLog("Found audio transceiver:", audioTransceiver ? {
direction: audioTransceiver.direction,
mid: audioTransceiver.mid,
senderTrack: audioTransceiver.sender.track?.kind,
@ -276,10 +277,10 @@ export function useMicrophone() {
// Use the existing audio transceiver's sender
await audioTransceiver.sender.replaceTrack(audioTrack);
sender = audioTransceiver.sender;
console.log("Replaced audio track on existing transceiver");
devLog("Replaced audio track on existing transceiver");
// Verify the track was set correctly
console.log("Transceiver after track replacement:", {
devLog("Transceiver after track replacement:", {
direction: audioTransceiver.direction,
senderTrack: audioTransceiver.sender.track?.id,
senderTrackKind: audioTransceiver.sender.track?.kind,
@ -289,11 +290,11 @@ export function useMicrophone() {
} else {
// Fallback: add new track if no transceiver found
sender = peerConnection.addTrack(audioTrack, stream);
console.log("Added new audio track to peer connection");
devLog("Added new audio track to peer connection");
// Find the transceiver that was created for this track
const newTransceiver = peerConnection.getTransceivers().find(t => t.sender === sender);
console.log("New transceiver created:", newTransceiver ? {
devLog("New transceiver created:", newTransceiver ? {
direction: newTransceiver.direction,
senderTrack: newTransceiver.sender.track?.id,
senderTrackKind: newTransceiver.sender.track?.kind
@ -301,7 +302,7 @@ export function useMicrophone() {
}
setMicrophoneSender(sender);
console.log("Microphone sender set:", {
devLog("Microphone sender set:", {
senderId: sender,
track: sender.track?.id,
trackKind: sender.track?.kind,
@ -310,28 +311,30 @@ export function useMicrophone() {
});
// Check sender stats to verify audio is being transmitted
setTimeout(async () => {
try {
const stats = await sender.getStats();
console.log("Sender stats after 2 seconds:");
stats.forEach((report, id) => {
if (report.type === 'outbound-rtp' && report.kind === 'audio') {
console.log("Outbound audio RTP stats:", {
id,
packetsSent: report.packetsSent,
bytesSent: report.bytesSent,
timestamp: report.timestamp
});
}
});
} catch (error) {
console.error("Failed to get sender stats:", error);
}
}, 2000);
devOnly(() => {
setTimeout(async () => {
try {
const stats = await sender.getStats();
devLog("Sender stats after 2 seconds:");
stats.forEach((report, id) => {
if (report.type === 'outbound-rtp' && report.kind === 'audio') {
devLog("Outbound audio RTP stats:", {
id,
packetsSent: report.packetsSent,
bytesSent: report.bytesSent,
timestamp: report.timestamp
});
}
});
} catch (error) {
devError("Failed to get sender stats:", error);
}
}, 2000);
});
}
// Notify backend that microphone is started
console.log("Notifying backend about microphone start...");
devLog("Notifying backend about microphone start...");
// Retry logic for backend failures
let backendSuccess = false;
@ -341,12 +344,12 @@ export function useMicrophone() {
try {
// If this is a retry, first try to reset the backend microphone state
if (attempt > 1) {
console.log(`Backend start attempt ${attempt}, first trying to reset backend state...`);
devLog(`Backend start attempt ${attempt}, first trying to reset backend state...`);
try {
// Try the new reset endpoint first
const resetResp = await api.POST("/microphone/reset", {});
if (resetResp.ok) {
console.log("Backend reset successful");
devLog("Backend reset successful");
} else {
// Fallback to stop
await api.POST("/microphone/stop", {});
@ -354,59 +357,59 @@ export function useMicrophone() {
// Wait a bit for the backend to reset
await new Promise(resolve => setTimeout(resolve, 200));
} catch (resetError) {
console.warn("Failed to reset backend state:", resetError);
devWarn("Failed to reset backend state:", resetError);
}
}
const backendResp = await api.POST("/microphone/start", {});
console.log(`Backend response status (attempt ${attempt}):`, backendResp.status, "ok:", backendResp.ok);
devLog(`Backend response status (attempt ${attempt}):`, backendResp.status, "ok:", backendResp.ok);
if (!backendResp.ok) {
lastError = `Backend returned status ${backendResp.status}`;
console.error(`Backend microphone start failed with status: ${backendResp.status} (attempt ${attempt})`);
devError(`Backend microphone start failed with status: ${backendResp.status} (attempt ${attempt})`);
// For 500 errors, try again after a short delay
if (backendResp.status === 500 && attempt < 3) {
console.log(`Retrying backend start in 500ms (attempt ${attempt + 1}/3)...`);
devLog(`Retrying backend start in 500ms (attempt ${attempt + 1}/3)...`);
await new Promise(resolve => setTimeout(resolve, 500));
continue;
}
} else {
// Success!
const responseData = await backendResp.json();
console.log("Backend response data:", responseData);
devLog("Backend response data:", responseData);
if (responseData.status === "already running") {
console.info("Backend microphone was already running");
devInfo("Backend microphone was already running");
// If we're on the first attempt and backend says "already running",
// but frontend thinks it's not active, this might be a stuck state
if (attempt === 1 && !isMicrophoneActive) {
console.warn("Backend reports 'already running' but frontend is not active - possible stuck state");
console.log("Attempting to reset backend state and retry...");
devWarn("Backend reports 'already running' but frontend is not active - possible stuck state");
devLog("Attempting to reset backend state and retry...");
try {
const resetResp = await api.POST("/microphone/reset", {});
if (resetResp.ok) {
console.log("Backend reset successful, retrying start...");
devLog("Backend reset successful, retrying start...");
await new Promise(resolve => setTimeout(resolve, 200));
continue; // Retry the start
}
} catch (resetError) {
console.warn("Failed to reset stuck backend state:", resetError);
devWarn("Failed to reset stuck backend state:", resetError);
}
}
}
console.log("Backend microphone start successful");
devLog("Backend microphone start successful");
backendSuccess = true;
break;
}
} catch (error) {
lastError = error instanceof Error ? error : String(error);
console.error(`Backend microphone start threw error (attempt ${attempt}):`, error);
devError(`Backend microphone start threw error (attempt ${attempt}):`, error);
// For network errors, try again after a short delay
if (attempt < 3) {
console.log(`Retrying backend start in 500ms (attempt ${attempt + 1}/3)...`);
devLog(`Retrying backend start in 500ms (attempt ${attempt + 1}/3)...`);
await new Promise(resolve => setTimeout(resolve, 500));
continue;
}
@ -415,7 +418,7 @@ export function useMicrophone() {
// If all backend attempts failed, cleanup and return error
if (!backendSuccess) {
console.error("All backend start attempts failed, cleaning up stream");
devError("All backend start attempts failed, cleaning up stream");
await stopMicrophoneStream();
isStartingRef.current = false;
setIsStarting(false);
@ -432,7 +435,7 @@ export function useMicrophone() {
setMicrophoneActive(true);
setMicrophoneMuted(false);
console.log("Microphone state set to active. Verifying state:", {
devLog("Microphone state set to active. Verifying state:", {
streamInRef: !!microphoneStreamRef.current,
streamInStore: !!microphoneStream,
isActive: true,
@ -441,15 +444,17 @@ export function useMicrophone() {
// Don't sync immediately after starting - it causes race conditions
// The sync will happen naturally through other triggers
setTimeout(() => {
// Just verify state after a delay for debugging
console.log("State check after delay:", {
streamInRef: !!microphoneStreamRef.current,
streamInStore: !!microphoneStream,
isActive: isMicrophoneActive,
isMuted: isMicrophoneMuted
});
}, 100);
devOnly(() => {
setTimeout(() => {
// Just verify state after a delay for debugging
devLog("State check after delay:", {
streamInRef: !!microphoneStreamRef.current,
streamInStore: !!microphoneStream,
isActive: isMicrophoneActive,
isMuted: isMicrophoneMuted
});
}, AUDIO_CONFIG.AUDIO_TEST_TIMEOUT);
});
// Clear the starting flag
isStartingRef.current = false;
@ -493,12 +498,12 @@ export function useMicrophone() {
// Reset backend microphone state
const resetBackendMicrophoneState = useCallback(async (): Promise<boolean> => {
try {
console.log("Resetting backend microphone state...");
devLog("Resetting backend microphone state...");
const response = await api.POST("/microphone/reset", {});
if (response.ok) {
const data = await response.json();
console.log("Backend microphone reset successful:", data);
devLog("Backend microphone reset successful:", data);
// Update frontend state to match backend
setMicrophoneActive(false);
@ -506,7 +511,7 @@ export function useMicrophone() {
// Clean up any orphaned streams
if (microphoneStreamRef.current) {
console.log("Cleaning up orphaned stream after reset");
devLog("Cleaning up orphaned stream after reset");
await stopMicrophoneStream();
}
@ -518,19 +523,19 @@ export function useMicrophone() {
return true;
} else {
console.error("Backend microphone reset failed:", response.status);
devError("Backend microphone reset failed:", response.status);
return false;
}
} catch (error) {
console.warn("Failed to reset backend microphone state:", error);
devWarn("Failed to reset backend microphone state:", error);
// Fallback to old method
try {
console.log("Trying fallback reset method...");
devLog("Trying fallback reset method...");
await api.POST("/microphone/stop", {});
await new Promise(resolve => setTimeout(resolve, 300));
return true;
} catch (fallbackError) {
console.error("Fallback reset also failed:", fallbackError);
devError("Fallback reset also failed:", fallbackError);
return false;
}
}
@ -540,7 +545,7 @@ export function useMicrophone() {
const stopMicrophone = useCallback(async (): Promise<{ success: boolean; error?: MicrophoneError }> => {
// Prevent multiple simultaneous stop operations
if (isStarting || isStopping || isToggling) {
console.log("Microphone operation already in progress, skipping stop");
devLog("Microphone operation already in progress, skipping stop");
return { success: false, error: { type: 'unknown', message: 'Operation already in progress' } };
}
@ -552,9 +557,9 @@ export function useMicrophone() {
// Then notify backend that microphone is stopped
try {
await api.POST("/microphone/stop", {});
console.log("Backend notified about microphone stop");
devLog("Backend notified about microphone stop");
} catch (error) {
console.warn("Failed to notify backend about microphone stop:", error);
devWarn("Failed to notify backend about microphone stop:", error);
}
// Update frontend state immediately
@ -567,7 +572,7 @@ export function useMicrophone() {
setIsStopping(false);
return { success: true };
} catch (error) {
console.error("Failed to stop microphone:", error);
devError("Failed to stop microphone:", error);
setIsStopping(false);
return {
success: false,
@ -583,7 +588,7 @@ export function useMicrophone() {
const toggleMicrophoneMute = useCallback(async (): Promise<{ success: boolean; error?: MicrophoneError }> => {
// Prevent multiple simultaneous toggle operations
if (isStarting || isStopping || isToggling) {
console.log("Microphone operation already in progress, skipping toggle");
devLog("Microphone operation already in progress, skipping toggle");
return { success: false, error: { type: 'unknown', message: 'Operation already in progress' } };
}
@ -592,7 +597,7 @@ export function useMicrophone() {
// Use the ref instead of store value to avoid race conditions
const currentStream = microphoneStreamRef.current || microphoneStream;
console.log("Toggle microphone mute - current state:", {
devLog("Toggle microphone mute - current state:", {
hasRefStream: !!microphoneStreamRef.current,
hasStoreStream: !!microphoneStream,
isActive: isMicrophoneActive,
@ -610,7 +615,7 @@ export function useMicrophone() {
streamId: currentStream?.id,
audioTracks: currentStream?.getAudioTracks().length || 0
};
console.warn("Microphone mute failed: stream or active state missing", errorDetails);
devWarn("Microphone mute failed: stream or active state missing", errorDetails);
// Provide more specific error message
let errorMessage = 'Microphone is not active';
@ -647,7 +652,7 @@ export function useMicrophone() {
// Mute/unmute the audio track
audioTracks.forEach(track => {
track.enabled = !newMutedState;
console.log(`Audio track ${track.id} enabled: ${track.enabled}`);
devLog(`Audio track ${track.id} enabled: ${track.enabled}`);
});
setMicrophoneMuted(newMutedState);
@ -656,13 +661,13 @@ export function useMicrophone() {
try {
await api.POST("/microphone/mute", { muted: newMutedState });
} catch (error) {
console.warn("Failed to notify backend about microphone mute:", error);
devWarn("Failed to notify backend about microphone mute:", error);
}
setIsToggling(false);
return { success: true };
} catch (error) {
console.error("Failed to toggle microphone mute:", error);
devError("Failed to toggle microphone mute:", error);
setIsToggling(false);
return {
success: false,
@ -677,7 +682,7 @@ export function useMicrophone() {
// Function to check WebRTC audio transmission stats
const checkAudioTransmissionStats = useCallback(async () => {
if (!microphoneSender) {
console.log("No microphone sender available");
devLog("No microphone sender available");
return null;
}
@ -707,38 +712,38 @@ export function useMicrophone() {
}
});
console.log("Audio transmission stats:", audioStats);
devLog("Audio transmission stats:", audioStats);
return audioStats;
} catch (error) {
console.error("Failed to get audio transmission stats:", error);
devError("Failed to get audio transmission stats:", error);
return null;
}
}, [microphoneSender]);
// Comprehensive test function to diagnose microphone issues
const testMicrophoneAudio = useCallback(async () => {
console.log("=== MICROPHONE AUDIO TEST ===");
devLog("=== MICROPHONE AUDIO TEST ===");
// 1. Check if we have a stream
const stream = microphoneStreamRef.current;
if (!stream) {
console.log("❌ No microphone stream available");
devLog("❌ No microphone stream available");
return;
}
console.log("✅ Microphone stream exists:", stream.id);
devLog("✅ Microphone stream exists:", stream.id);
// 2. Check audio tracks
const audioTracks = stream.getAudioTracks();
console.log("Audio tracks:", audioTracks.length);
devLog("Audio tracks:", audioTracks.length);
if (audioTracks.length === 0) {
console.log("❌ No audio tracks in stream");
devLog("❌ No audio tracks in stream");
return;
}
const track = audioTracks[0];
console.log("✅ Audio track details:", {
devLog("✅ Audio track details:", {
id: track.id,
label: track.label,
enabled: track.enabled,
@ -752,13 +757,13 @@ export function useMicrophone() {
const analyser = audioContext.createAnalyser();
const source = audioContext.createMediaStreamSource(stream);
analyser.fftSize = 256;
analyser.fftSize = AUDIO_CONFIG.ANALYSIS_FFT_SIZE;
source.connect(analyser);
const dataArray = new Uint8Array(analyser.frequencyBinCount);
console.log("🎤 Testing audio level detection for 5 seconds...");
console.log("Please speak into your microphone now!");
devLog("🎤 Testing audio level detection for 5 seconds...");
devLog("Please speak into your microphone now!");
let maxLevel = 0;
let sampleCount = 0;
@ -771,39 +776,39 @@ export function useMicrophone() {
sum += value * value;
}
const rms = Math.sqrt(sum / dataArray.length);
const level = Math.min(100, (rms / 255) * 100);
const level = Math.min(AUDIO_CONFIG.MAX_LEVEL_PERCENTAGE, (rms / AUDIO_CONFIG.LEVEL_SCALING_FACTOR) * AUDIO_CONFIG.MAX_LEVEL_PERCENTAGE);
maxLevel = Math.max(maxLevel, level);
sampleCount++;
if (sampleCount % 10 === 0) { // Log every 10th sample
console.log(`Audio level: ${level.toFixed(1)}% (max so far: ${maxLevel.toFixed(1)}%)`);
devLog(`Audio level: ${level.toFixed(1)}% (max so far: ${maxLevel.toFixed(1)}%)`);
}
}, 100);
}, AUDIO_CONFIG.ANALYSIS_UPDATE_INTERVAL);
setTimeout(() => {
clearInterval(testInterval);
source.disconnect();
audioContext.close();
console.log("🎤 Audio test completed!");
console.log(`Maximum audio level detected: ${maxLevel.toFixed(1)}%`);
devLog("🎤 Audio test completed!");
devLog(`Maximum audio level detected: ${maxLevel.toFixed(1)}%`);
if (maxLevel > 5) {
console.log("✅ Microphone is detecting audio!");
devLog("✅ Microphone is detecting audio!");
} else {
console.log("❌ No significant audio detected. Check microphone permissions and hardware.");
devLog("❌ No significant audio detected. Check microphone permissions and hardware.");
}
}, 5000);
}, NETWORK_CONFIG.AUDIO_TEST_DURATION);
} catch (error) {
console.error("❌ Failed to test audio level:", error);
devError("❌ Failed to test audio level:", error);
}
// 4. Check WebRTC sender
if (microphoneSender) {
console.log("✅ WebRTC sender exists");
console.log("Sender track:", {
devLog("✅ WebRTC sender exists");
devLog("Sender track:", {
id: microphoneSender.track?.id,
kind: microphoneSender.track?.kind,
enabled: microphoneSender.track?.enabled,
@ -812,45 +817,45 @@ export function useMicrophone() {
// Check if sender track matches stream track
if (microphoneSender.track === track) {
console.log("✅ Sender track matches stream track");
devLog("✅ Sender track matches stream track");
} else {
console.log("❌ Sender track does NOT match stream track");
devLog("❌ Sender track does NOT match stream track");
}
} else {
console.log("❌ No WebRTC sender available");
devLog("❌ No WebRTC sender available");
}
// 5. Check peer connection
if (peerConnection) {
console.log("✅ Peer connection exists");
console.log("Connection state:", peerConnection.connectionState);
console.log("ICE connection state:", peerConnection.iceConnectionState);
devLog("✅ Peer connection exists");
devLog("Connection state:", peerConnection.connectionState);
devLog("ICE connection state:", peerConnection.iceConnectionState);
const transceivers = peerConnection.getTransceivers();
const audioTransceivers = transceivers.filter(t =>
t.sender.track?.kind === 'audio' || t.receiver.track?.kind === 'audio'
);
console.log("Audio transceivers:", audioTransceivers.map(t => ({
devLog("Audio transceivers:", audioTransceivers.map(t => ({
direction: t.direction,
senderTrack: t.sender.track?.id,
receiverTrack: t.receiver.track?.id
})));
} else {
console.log("❌ No peer connection available");
devLog("❌ No peer connection available");
}
}, [microphoneSender, peerConnection]);
const startMicrophoneDebounced = useCallback((deviceId?: string) => {
debouncedOperation(async () => {
await startMicrophone(deviceId).catch(console.error);
await startMicrophone(deviceId).catch(devError);
}, "start");
}, [startMicrophone, debouncedOperation]);
const stopMicrophoneDebounced = useCallback(() => {
debouncedOperation(async () => {
await stopMicrophone().catch(console.error);
await stopMicrophone().catch(devError);
}, "stop");
}, [stopMicrophone, debouncedOperation]);
@ -919,10 +924,10 @@ export function useMicrophone() {
// Clean up stream directly without depending on the callback
const stream = microphoneStreamRef.current;
if (stream) {
console.log("Cleanup: stopping microphone stream on unmount");
devLog("Cleanup: stopping microphone stream on unmount");
stream.getAudioTracks().forEach(track => {
track.stop();
console.log(`Cleanup: stopped audio track ${track.id}`);
devLog(`Cleanup: stopped audio track ${track.id}`);
});
microphoneStreamRef.current = null;
}

View File

@ -1,5 +1,7 @@
import { useCallback, useEffect, useState } from "react";
import { devError } from '../utils/debug';
import { JsonRpcResponse, useJsonRpc } from "./useJsonRpc";
import { useAudioEvents } from "./useAudioEvents";
@ -25,7 +27,7 @@ export function useUsbDeviceConfig() {
setLoading(false);
if ("error" in resp) {
console.error("Failed to load USB devices:", resp.error);
devError("Failed to load USB devices:", resp.error);
setError(resp.error.data || "Unknown error");
setUsbDeviceConfig(null);
} else {

View File

@ -0,0 +1,142 @@
import api from '@/api';
interface AudioConfig {
Quality: number;
Bitrate: number;
SampleRate: number;
Channels: number;
FrameSize: string;
}
type QualityPresets = Record<number, AudioConfig>;
interface AudioQualityResponse {
current: AudioConfig;
presets: QualityPresets;
}
class AudioQualityService {
private audioPresets: QualityPresets | null = null;
private microphonePresets: QualityPresets | null = null;
private qualityLabels: Record<number, string> = {
0: 'Low',
1: 'Medium',
2: 'High',
3: 'Ultra'
};
/**
* Fetch audio quality presets from the backend
*/
async fetchAudioQualityPresets(): Promise<AudioQualityResponse | null> {
try {
const response = await api.GET('/audio/quality');
if (response.ok) {
const data = await response.json();
this.audioPresets = data.presets;
this.updateQualityLabels(data.presets);
return data;
}
} catch (error) {
console.error('Failed to fetch audio quality presets:', error);
}
return null;
}
/**
* Fetch microphone quality presets from the backend
*/
async fetchMicrophoneQualityPresets(): Promise<AudioQualityResponse | null> {
try {
const response = await api.GET('/microphone/quality');
if (response.ok) {
const data = await response.json();
this.microphonePresets = data.presets;
return data;
}
} catch (error) {
console.error('Failed to fetch microphone quality presets:', error);
}
return null;
}
/**
* Update quality labels with actual bitrates from presets
*/
private updateQualityLabels(presets: QualityPresets): void {
const newQualityLabels: Record<number, string> = {};
Object.entries(presets).forEach(([qualityNum, preset]) => {
const quality = parseInt(qualityNum);
const qualityNames = ['Low', 'Medium', 'High', 'Ultra'];
const name = qualityNames[quality] || `Quality ${quality}`;
newQualityLabels[quality] = `${name} (${preset.Bitrate}kbps)`;
});
this.qualityLabels = newQualityLabels;
}
/**
* Get quality labels with bitrates
*/
getQualityLabels(): Record<number, string> {
return this.qualityLabels;
}
/**
* Get cached audio presets
*/
getAudioPresets(): QualityPresets | null {
return this.audioPresets;
}
/**
* Get cached microphone presets
*/
getMicrophonePresets(): QualityPresets | null {
return this.microphonePresets;
}
/**
* Set audio quality
*/
async setAudioQuality(quality: number): Promise<boolean> {
try {
const response = await api.POST('/audio/quality', { quality });
return response.ok;
} catch (error) {
console.error('Failed to set audio quality:', error);
return false;
}
}
/**
* Set microphone quality
*/
async setMicrophoneQuality(quality: number): Promise<boolean> {
try {
const response = await api.POST('/microphone/quality', { quality });
return response.ok;
} catch (error) {
console.error('Failed to set microphone quality:', error);
return false;
}
}
/**
* Load both audio and microphone configurations
*/
async loadAllConfigurations(): Promise<{
audio: AudioQualityResponse | null;
microphone: AudioQualityResponse | null;
}> {
const [audio, microphone] = await Promise.all([
this.fetchAudioQualityPresets(),
this.fetchMicrophoneQualityPresets()
]);
return { audio, microphone };
}
}
// Export a singleton instance
export const audioQualityService = new AudioQualityService();
export default audioQualityService;

64
ui/src/utils/debug.ts Normal file
View File

@ -0,0 +1,64 @@
/**
* Debug utilities for development mode logging
*/
// Check if we're in development mode
const isDevelopment = import.meta.env.DEV || import.meta.env.MODE === 'development';
/**
* Development-only console.log wrapper
* Only logs in development mode, silent in production
*/
export const devLog = (...args: unknown[]): void => {
if (isDevelopment) {
console.log(...args);
}
};
/**
* Development-only console.info wrapper
* Only logs in development mode, silent in production
*/
export const devInfo = (...args: unknown[]): void => {
if (isDevelopment) {
console.info(...args);
}
};
/**
* Development-only console.warn wrapper
* Only logs in development mode, silent in production
*/
export const devWarn = (...args: unknown[]): void => {
if (isDevelopment) {
console.warn(...args);
}
};
/**
* Development-only console.error wrapper
* Always logs errors, but with dev prefix in development
*/
export const devError = (...args: unknown[]): void => {
if (isDevelopment) {
console.error('[DEV]', ...args);
} else {
console.error(...args);
}
};
/**
* Development-only debug function wrapper
* Only executes the function in development mode
*/
export const devOnly = <T>(fn: () => T): T | undefined => {
if (isDevelopment) {
return fn();
}
return undefined;
};
/**
* Check if we're in development mode
*/
export const isDevMode = (): boolean => isDevelopment;