mirror/tracy
1
0
Fork 0
mirror of https://github.com/wolfpld/tracy.git synced 2024-09-20 13:52:18 +00:00
Code Issues Packages Projects Releases Wiki Activity
5177629130
tracy/server/TracyWorker.hpp

520 lines
21 KiB
C++
Raw Normal View History

Split data handling code from the view.
2018-02-13 13:57:47 +00:00
#ifndef __TRACYWORKER_HPP__
#define __TRACYWORKER_HPP__
#include <atomic>
Calculate min/max time spans for source locations.
2018-03-18 19:15:45 +00:00
#include <limits>
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
#include <map>
Throw exception when trying to open unsupported dump version.
2018-04-21 12:18:13 +00:00
#include <stdexcept>
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
#include <string>
#include <thread>
#include <vector>
#include "../common/tracy_lz4.hpp"
#include "../common/TracyForceInline.hpp"
Use the fastest mutex available. The selection is based on the following test results: MSVC: === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 11.641 ns/iter 2 thread contention: 141.559 ns/iter 3 thread contention: 242.733 ns/iter 4 thread contention: 409.807 ns/iter 5 thread contention: 561.544 ns/iter 6 thread contention: 785.845 ns/iter => std::mutex No contention: 19.190 ns/iter 2 thread contention: 39.305 ns/iter 3 thread contention: 58.999 ns/iter 4 thread contention: 59.532 ns/iter 5 thread contention: 103.539 ns/iter 6 thread contention: 110.314 ns/iter => std::shared_timed_mutex No contention: 45.487 ns/iter 2 thread contention: 96.351 ns/iter 3 thread contention: 142.871 ns/iter 4 thread contention: 184.999 ns/iter 5 thread contention: 336.608 ns/iter 6 thread contention: 542.551 ns/iter => std::shared_mutex No contention: 10.861 ns/iter 2 thread contention: 17.495 ns/iter 3 thread contention: 31.126 ns/iter 4 thread contention: 40.468 ns/iter 5 thread contention: 15.677 ns/iter 6 thread contention: 64.505 ns/iter Cygwin (clang): === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 11.536 ns/iter 2 thread contention: 121.082 ns/iter 3 thread contention: 396.430 ns/iter 4 thread contention: 672.555 ns/iter 5 thread contention: 1327.761 ns/iter 6 thread contention: 14151.955 ns/iter => std::mutex No contention: 62.583 ns/iter 2 thread contention: 3990.464 ns/iter 3 thread contention: 7161.189 ns/iter 4 thread contention: 9870.820 ns/iter 5 thread contention: 12355.178 ns/iter 6 thread contention: 14694.903 ns/iter => std::shared_timed_mutex No contention: 91.687 ns/iter 2 thread contention: 1115.037 ns/iter 3 thread contention: 4183.792 ns/iter 4 thread contention: 15283.491 ns/iter 5 thread contention: 27812.477 ns/iter 6 thread contention: 35028.140 ns/iter => std::shared_mutex No contention: 91.764 ns/iter 2 thread contention: 1051.826 ns/iter 3 thread contention: 5574.720 ns/iter 4 thread contention: 15721.416 ns/iter 5 thread contention: 27721.487 ns/iter 6 thread contention: 35420.404 ns/iter Linux (x64): === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 13.487 ns/iter 2 thread contention: 210.317 ns/iter 3 thread contention: 430.855 ns/iter 4 thread contention: 510.533 ns/iter 5 thread contention: 1003.609 ns/iter 6 thread contention: 1787.683 ns/iter => std::mutex No contention: 12.403 ns/iter 2 thread contention: 157.122 ns/iter 3 thread contention: 186.791 ns/iter 4 thread contention: 265.073 ns/iter 5 thread contention: 283.778 ns/iter 6 thread contention: 270.687 ns/iter => std::shared_timed_mutex No contention: 21.509 ns/iter 2 thread contention: 150.179 ns/iter 3 thread contention: 256.574 ns/iter 4 thread contention: 415.351 ns/iter 5 thread contention: 611.532 ns/iter 6 thread contention: 944.695 ns/iter => std::shared_mutex No contention: 20.805 ns/iter 2 thread contention: 157.034 ns/iter 3 thread contention: 244.025 ns/iter 4 thread contention: 406.269 ns/iter 5 thread contention: 387.985 ns/iter 6 thread contention: 468.550 ns/iter Linux (arm64): === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 20.891 ns/iter 2 thread contention: 211.037 ns/iter 3 thread contention: 409.962 ns/iter 4 thread contention: 657.441 ns/iter 5 thread contention: 828.405 ns/iter 6 thread contention: 1131.827 ns/iter => std::mutex No contention: 50.884 ns/iter 2 thread contention: 103.620 ns/iter 3 thread contention: 332.429 ns/iter 4 thread contention: 620.802 ns/iter 5 thread contention: 783.943 ns/iter 6 thread contention: 834.002 ns/iter => std::shared_timed_mutex No contention: 64.948 ns/iter 2 thread contention: 173.191 ns/iter 3 thread contention: 490.352 ns/iter 4 thread contention: 660.668 ns/iter 5 thread contention: 1014.546 ns/iter 6 thread contention: 1451.553 ns/iter => std::shared_mutex No contention: 64.521 ns/iter 2 thread contention: 195.222 ns/iter 3 thread contention: 490.819 ns/iter 4 thread contention: 654.786 ns/iter 5 thread contention: 955.759 ns/iter 6 thread contention: 1282.544 ns/iter
2018-07-13 22:39:01 +00:00
#include "../common/TracyMutex.hpp"
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
#include "../common/TracyQueue.hpp"
#include "../common/TracySocket.hpp"
#include "tracy_flat_hash_map.hpp"
#include "TracyEvent.hpp"
#include "TracySlab.hpp"
Use StringDiscovery for plots.
2018-08-04 14:33:03 +00:00
#include "TracyStringDiscovery.hpp"
Store unique callstack payloads.
2018-06-19 19:15:36 +00:00
#include "TracyVarArray.hpp"
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
namespace tracy
{
class FileRead;
class FileWrite;
Allow partial load of data from dump.
2018-04-20 14:03:09 +00:00
namespace EventType
{
enum Type : uint32_t
{
Locks = 1 << 0,
Messages = 1 << 1,
Plots = 1 << 2,
Memory = 1 << 3,
None = 0,
All = std::numeric_limits<uint32_t>::max()
};
}
Throw exception when trying to open unsupported dump version.
2018-04-21 12:18:13 +00:00
struct UnsupportedVersion : public std::exception
{
UnsupportedVersion( int version ) : version( version ) {}
int version;
};
Track trace loading progress.
2018-07-28 15:59:17 +00:00
struct LoadProgress
{
Display explicit names of loaded things.
2018-07-29 14:56:46 +00:00
enum Stage
{
Initialization,
Locks,
Messages,
Zones,
GpuZones,
Plots,
Memory,
CallStacks
};
Fix initialization of atomics.
2018-07-28 18:13:06 +00:00
LoadProgress() : total( 0 ), progress( 0 ), subTotal( 0 ), subProgress( 0 ) {}
std::atomic<uint64_t> total;
std::atomic<uint64_t> progress;
std::atomic<uint64_t> subTotal;
std::atomic<uint64_t> subProgress;
Track trace loading progress.
2018-07-28 15:59:17 +00:00
};
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
class Worker
{
Zones can be now also grouped by call stack.
2018-07-21 18:26:13 +00:00
public:
Store thread id next to zone ptr in source location zone list.
2018-03-18 19:45:49 +00:00
#pragma pack( 1 )
struct ZoneThreadData
{
ZoneEvent* zone;
uint16_t thread;
};
#pragma pack()
Zones can be now also grouped by call stack.
2018-07-21 18:26:13 +00:00
private:
Put source location zone data into a struct.
2018-03-18 19:08:57 +00:00
struct SourceLocationZones
{
Store thread id next to zone ptr in source location zone list.
2018-03-18 19:45:49 +00:00
Vector<ZoneThreadData> zones;
Initialize SourceLocationZones in-place.
2019-03-14 00:15:19 +00:00
int64_t min = std::numeric_limits<int64_t>::max();
int64_t max = std::numeric_limits<int64_t>::min();
int64_t total = 0;
Optimize calculation of standard deviation.
2019-03-14 00:23:37 +00:00
double sumSq = 0;
Initialize SourceLocationZones in-place.
2019-03-14 00:15:19 +00:00
int64_t selfMin = std::numeric_limits<int64_t>::max();
int64_t selfMax = std::numeric_limits<int64_t>::min();
int64_t selfTotal = 0;
Put source location zone data into a struct.
2018-03-18 19:08:57 +00:00
};
Add callstack frame identifier and the required plumbing.
2019-03-03 15:37:21 +00:00
struct CallstackFrameIdHash
{
Add missing const qualifiers.
2019-03-03 16:54:00 +00:00
size_t operator()( const CallstackFrameId& id ) const { return id.data; }
Add callstack frame identifier and the required plumbing.
2019-03-03 15:37:21 +00:00
typedef tracy::power_of_two_hash_policy hash_policy;
};
struct CallstackFrameIdCompare
{
Add missing const qualifiers.
2019-03-03 16:54:00 +00:00
bool operator()( const CallstackFrameId& lhs, const CallstackFrameId& rhs ) const { return lhs.data == rhs.data; }
Add callstack frame identifier and the required plumbing.
2019-03-03 15:37:21 +00:00
};
Store callstack alloc payloads.
2019-03-03 16:34:56 +00:00
struct RevFrameHash
{
Add missing const qualifiers.
2019-03-03 16:54:00 +00:00
size_t operator()( const CallstackFrameData* data ) const
Store callstack alloc payloads.
2019-03-03 16:34:56 +00:00
{
size_t hash = data->size;
for( uint8_t i=0; i<data->size; i++ )
{
const auto& v = data->data[i];
hash = ( ( hash << 5 ) + hash ) ^ size_t( v.line );
hash = ( ( hash << 5 ) + hash ) ^ size_t( v.file.__data );
hash = ( ( hash << 5 ) + hash ) ^ size_t( v.name.__data );
}
return hash;
}
typedef tracy::power_of_two_hash_policy hash_policy;
};
struct RevFrameComp
{
Add missing const qualifiers.
2019-03-03 16:54:00 +00:00
bool operator()( const CallstackFrameData* lhs, const CallstackFrameData* rhs ) const
Store callstack alloc payloads.
2019-03-03 16:34:56 +00:00
{
if( lhs->size != rhs->size ) return false;
for( uint8_t i=0; i<lhs->size; i++ )
{
if( memcmp( lhs->data + i, rhs->data + i, sizeof( CallstackFrame ) ) != 0 ) return false;
}
return true;
}
};
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
struct DataBlock
{
Store on-demand frame offset.
2018-07-10 20:39:41 +00:00
DataBlock() : zonesCnt( 0 ), lastTime( 0 ), frameOffset( 0 ), threadLast( std::numeric_limits<uint64_t>::max(), 0 ) {}
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
Use the fastest mutex available. The selection is based on the following test results: MSVC: === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 11.641 ns/iter 2 thread contention: 141.559 ns/iter 3 thread contention: 242.733 ns/iter 4 thread contention: 409.807 ns/iter 5 thread contention: 561.544 ns/iter 6 thread contention: 785.845 ns/iter => std::mutex No contention: 19.190 ns/iter 2 thread contention: 39.305 ns/iter 3 thread contention: 58.999 ns/iter 4 thread contention: 59.532 ns/iter 5 thread contention: 103.539 ns/iter 6 thread contention: 110.314 ns/iter => std::shared_timed_mutex No contention: 45.487 ns/iter 2 thread contention: 96.351 ns/iter 3 thread contention: 142.871 ns/iter 4 thread contention: 184.999 ns/iter 5 thread contention: 336.608 ns/iter 6 thread contention: 542.551 ns/iter => std::shared_mutex No contention: 10.861 ns/iter 2 thread contention: 17.495 ns/iter 3 thread contention: 31.126 ns/iter 4 thread contention: 40.468 ns/iter 5 thread contention: 15.677 ns/iter 6 thread contention: 64.505 ns/iter Cygwin (clang): === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 11.536 ns/iter 2 thread contention: 121.082 ns/iter 3 thread contention: 396.430 ns/iter 4 thread contention: 672.555 ns/iter 5 thread contention: 1327.761 ns/iter 6 thread contention: 14151.955 ns/iter => std::mutex No contention: 62.583 ns/iter 2 thread contention: 3990.464 ns/iter 3 thread contention: 7161.189 ns/iter 4 thread contention: 9870.820 ns/iter 5 thread contention: 12355.178 ns/iter 6 thread contention: 14694.903 ns/iter => std::shared_timed_mutex No contention: 91.687 ns/iter 2 thread contention: 1115.037 ns/iter 3 thread contention: 4183.792 ns/iter 4 thread contention: 15283.491 ns/iter 5 thread contention: 27812.477 ns/iter 6 thread contention: 35028.140 ns/iter => std::shared_mutex No contention: 91.764 ns/iter 2 thread contention: 1051.826 ns/iter 3 thread contention: 5574.720 ns/iter 4 thread contention: 15721.416 ns/iter 5 thread contention: 27721.487 ns/iter 6 thread contention: 35420.404 ns/iter Linux (x64): === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 13.487 ns/iter 2 thread contention: 210.317 ns/iter 3 thread contention: 430.855 ns/iter 4 thread contention: 510.533 ns/iter 5 thread contention: 1003.609 ns/iter 6 thread contention: 1787.683 ns/iter => std::mutex No contention: 12.403 ns/iter 2 thread contention: 157.122 ns/iter 3 thread contention: 186.791 ns/iter 4 thread contention: 265.073 ns/iter 5 thread contention: 283.778 ns/iter 6 thread contention: 270.687 ns/iter => std::shared_timed_mutex No contention: 21.509 ns/iter 2 thread contention: 150.179 ns/iter 3 thread contention: 256.574 ns/iter 4 thread contention: 415.351 ns/iter 5 thread contention: 611.532 ns/iter 6 thread contention: 944.695 ns/iter => std::shared_mutex No contention: 20.805 ns/iter 2 thread contention: 157.034 ns/iter 3 thread contention: 244.025 ns/iter 4 thread contention: 406.269 ns/iter 5 thread contention: 387.985 ns/iter 6 thread contention: 468.550 ns/iter Linux (arm64): === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 20.891 ns/iter 2 thread contention: 211.037 ns/iter 3 thread contention: 409.962 ns/iter 4 thread contention: 657.441 ns/iter 5 thread contention: 828.405 ns/iter 6 thread contention: 1131.827 ns/iter => std::mutex No contention: 50.884 ns/iter 2 thread contention: 103.620 ns/iter 3 thread contention: 332.429 ns/iter 4 thread contention: 620.802 ns/iter 5 thread contention: 783.943 ns/iter 6 thread contention: 834.002 ns/iter => std::shared_timed_mutex No contention: 64.948 ns/iter 2 thread contention: 173.191 ns/iter 3 thread contention: 490.352 ns/iter 4 thread contention: 660.668 ns/iter 5 thread contention: 1014.546 ns/iter 6 thread contention: 1451.553 ns/iter => std::shared_mutex No contention: 64.521 ns/iter 2 thread contention: 195.222 ns/iter 3 thread contention: 490.819 ns/iter 4 thread contention: 654.786 ns/iter 5 thread contention: 955.759 ns/iter 6 thread contention: 1282.544 ns/iter
2018-07-13 22:39:01 +00:00
TracyMutex lock;
Multiple frame sets support.
2018-08-04 17:47:09 +00:00
StringDiscovery<FrameData*> frames;
FrameData* framesBase;
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
Vector<GpuCtxData*> gpuData;
Vector<MessageData*> messages;
Use StringDiscovery for plots.
2018-08-04 14:33:03 +00:00
StringDiscovery<PlotData*> plots;
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
Vector<ThreadData*> threads;
Broken memory events processing.
2018-04-01 00:03:34 +00:00
MemData memory;
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
uint64_t zonesCnt;
int64_t lastTime;
Store on-demand frame offset.
2018-07-10 20:39:41 +00:00
uint64_t frameOffset;
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
flat_hash_map<uint64_t, const char*, nohash<uint64_t>> strings;
Vector<const char*> stringData;
Preserve string length in string map.
2019-02-12 19:23:14 +00:00
flat_hash_map<charutil::StringKey, uint32_t, charutil::StringKey::HasherPOT, charutil::StringKey::Comparator> stringMap;
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
flat_hash_map<uint64_t, const char*, nohash<uint64_t>> threadNames;
flat_hash_map<uint64_t, SourceLocation, nohash<uint64_t>> sourceLocation;
Vector<SourceLocation*> sourceLocationPayload;
flat_hash_map<SourceLocation*, uint32_t, SourceLocationHasher, SourceLocationComparator> sourceLocationPayloadMap;
Use Vector for source location expand storage.
2018-03-04 15:32:51 +00:00
Vector<uint64_t> sourceLocationExpand;
Reduce data collection if TRACY_NO_STATISTICS is defined. Statistical data collection is only useful if it's meant to be used. Otherwise it only incurs CPU and memory cost.
2018-03-18 11:55:54 +00:00
#ifndef TRACY_NO_STATISTICS
Put source location zone data into a struct.
2018-03-18 19:08:57 +00:00
flat_hash_map<int32_t, SourceLocationZones, nohash<int32_t>> sourceLocationZones;
Sort source location zones on a separate thread.
2018-04-30 01:54:09 +00:00
bool sourceLocationZonesReady;
Always count source location zones.
2018-07-29 12:16:13 +00:00
#else
flat_hash_map<int32_t, uint64_t> sourceLocationZonesCnt;
Reduce data collection if TRACY_NO_STATISTICS is defined. Statistical data collection is only useful if it's meant to be used. Otherwise it only incurs CPU and memory cost.
2018-03-18 11:55:54 +00:00
#endif
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
Replace frame pointers with callstack frame ids.
2019-03-03 15:50:18 +00:00
flat_hash_map<VarArray<CallstackFrameId>*, uint32_t, VarArrayHasherPOT<CallstackFrameId>, VarArrayComparator<CallstackFrameId>> callstackMap;
Vector<VarArray<CallstackFrameId>*> callstackPayload;
flat_hash_map<CallstackFrameId, CallstackFrameData*, CallstackFrameIdHash, CallstackFrameIdCompare> callstackFrameMap;
Store callstack alloc payloads.
2019-03-03 16:34:56 +00:00
flat_hash_map<CallstackFrameData*, CallstackFrameId, RevFrameHash, RevFrameComp> revFrameMap;
Store unique callstack payloads.
2018-06-19 19:15:36 +00:00
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
std::map<uint32_t, LockMap> lockMap;
Thread id compression/decompression.
2018-03-18 19:45:22 +00:00
flat_hash_map<uint64_t, uint16_t, nohash<uint64_t>> threadMap;
Vector<uint64_t> threadExpand;
Add thread compression cache. Observation: calls to CompressThread() are likely to be repeated with the same value. Exploit that by storing last query and its result.
2018-04-30 23:29:25 +00:00
std::pair<uint64_t, uint16_t> threadLast;
Store children vectors in a separate data collection. This reduces per-zone memory cost by 9 bytes if there are no children and increases it by 4 bytes, if there are children. This is universally a better solution, as the following data shows: +++ /home/wolf/desktop/tracy-old/android.tracy +++ Vectors: 2794480 Size 0: 2373070 (84.92%) Size 1: 70237 (2.51%) Size 2+: 351173 (12.57%) +++ /home/wolf/desktop/tracy-old/asset-new.tracy +++ Vectors: 1799227 Size 0: 1482691 (82.41%) Size 1: 93272 (5.18%) Size 2+: 223264 (12.41%) +++ /home/wolf/desktop/tracy-old/asset-new-id.tracy +++ Vectors: 1977996 Size 0: 1640817 (82.95%) Size 1: 97198 (4.91%) Size 2+: 239981 (12.13%) +++ /home/wolf/desktop/tracy-old/asset-old.tracy +++ Vectors: 1782395 Size 0: 1471437 (82.55%) Size 1: 88813 (4.98%) Size 2+: 222145 (12.46%) +++ /home/wolf/desktop/tracy-old/big.tracy +++ Vectors: 180794047 Size 0: 172696094 (95.52%) Size 1: 2799772 (1.55%) Size 2+: 5298181 (2.93%) +++ /home/wolf/desktop/tracy-old/darkrl.tracy +++ Vectors: 12014129 Size 0: 11611324 (96.65%) Size 1: 134980 (1.12%) Size 2+: 267825 (2.23%) +++ /home/wolf/desktop/tracy-old/mem.tracy +++ Vectors: 383097 Size 0: 321932 (84.03%) Size 1: 854 (0.22%) Size 2+: 60311 (15.74%) +++ /home/wolf/desktop/tracy-old/new.tracy +++ Vectors: 77536 Size 0: 63035 (81.30%) Size 1: 8886 (11.46%) Size 2+: 5615 (7.24%) +++ /home/wolf/desktop/tracy-old/selfprofile.tracy +++ Vectors: 22940871 Size 0: 22704868 (98.97%) Size 1: 73000 (0.32%) Size 2+: 163003 (0.71%) +++ /home/wolf/desktop/tracy-old/tbrowser.tracy +++ Vectors: 962682 Size 0: 695380 (72.23%) Size 1: 43007 (4.47%) Size 2+: 224295 (23.30%) +++ /home/wolf/desktop/tracy-old/virtualfile_hc.tracy +++ Vectors: 529170 Size 0: 449386 (84.92%) Size 1: 15694 (2.97%) Size 2+: 64090 (12.11%) +++ /home/wolf/desktop/tracy-old/zfile_hc.tracy +++ Vectors: 264849 Size 0: 220589 (83.29%) Size 1: 9386 (3.54%) Size 2+: 34874 (13.17%)
2018-07-22 14:05:50 +00:00
Use Vector to store children data instead of std::vector.
2019-02-13 01:32:25 +00:00
Vector<Vector<ZoneEvent*>> m_zoneChildren;
Vector<Vector<GpuEvent*>> m_gpuChildren;
Send crash report.
2018-08-20 00:07:31 +00:00
CrashEvent m_crashEvent;
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
};
struct MbpsBlock
{
MbpsBlock() : mbps( 64 ), compRatio( 1.0 ) {}
Use the fastest mutex available. The selection is based on the following test results: MSVC: === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 11.641 ns/iter 2 thread contention: 141.559 ns/iter 3 thread contention: 242.733 ns/iter 4 thread contention: 409.807 ns/iter 5 thread contention: 561.544 ns/iter 6 thread contention: 785.845 ns/iter => std::mutex No contention: 19.190 ns/iter 2 thread contention: 39.305 ns/iter 3 thread contention: 58.999 ns/iter 4 thread contention: 59.532 ns/iter 5 thread contention: 103.539 ns/iter 6 thread contention: 110.314 ns/iter => std::shared_timed_mutex No contention: 45.487 ns/iter 2 thread contention: 96.351 ns/iter 3 thread contention: 142.871 ns/iter 4 thread contention: 184.999 ns/iter 5 thread contention: 336.608 ns/iter 6 thread contention: 542.551 ns/iter => std::shared_mutex No contention: 10.861 ns/iter 2 thread contention: 17.495 ns/iter 3 thread contention: 31.126 ns/iter 4 thread contention: 40.468 ns/iter 5 thread contention: 15.677 ns/iter 6 thread contention: 64.505 ns/iter Cygwin (clang): === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 11.536 ns/iter 2 thread contention: 121.082 ns/iter 3 thread contention: 396.430 ns/iter 4 thread contention: 672.555 ns/iter 5 thread contention: 1327.761 ns/iter 6 thread contention: 14151.955 ns/iter => std::mutex No contention: 62.583 ns/iter 2 thread contention: 3990.464 ns/iter 3 thread contention: 7161.189 ns/iter 4 thread contention: 9870.820 ns/iter 5 thread contention: 12355.178 ns/iter 6 thread contention: 14694.903 ns/iter => std::shared_timed_mutex No contention: 91.687 ns/iter 2 thread contention: 1115.037 ns/iter 3 thread contention: 4183.792 ns/iter 4 thread contention: 15283.491 ns/iter 5 thread contention: 27812.477 ns/iter 6 thread contention: 35028.140 ns/iter => std::shared_mutex No contention: 91.764 ns/iter 2 thread contention: 1051.826 ns/iter 3 thread contention: 5574.720 ns/iter 4 thread contention: 15721.416 ns/iter 5 thread contention: 27721.487 ns/iter 6 thread contention: 35420.404 ns/iter Linux (x64): === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 13.487 ns/iter 2 thread contention: 210.317 ns/iter 3 thread contention: 430.855 ns/iter 4 thread contention: 510.533 ns/iter 5 thread contention: 1003.609 ns/iter 6 thread contention: 1787.683 ns/iter => std::mutex No contention: 12.403 ns/iter 2 thread contention: 157.122 ns/iter 3 thread contention: 186.791 ns/iter 4 thread contention: 265.073 ns/iter 5 thread contention: 283.778 ns/iter 6 thread contention: 270.687 ns/iter => std::shared_timed_mutex No contention: 21.509 ns/iter 2 thread contention: 150.179 ns/iter 3 thread contention: 256.574 ns/iter 4 thread contention: 415.351 ns/iter 5 thread contention: 611.532 ns/iter 6 thread contention: 944.695 ns/iter => std::shared_mutex No contention: 20.805 ns/iter 2 thread contention: 157.034 ns/iter 3 thread contention: 244.025 ns/iter 4 thread contention: 406.269 ns/iter 5 thread contention: 387.985 ns/iter 6 thread contention: 468.550 ns/iter Linux (arm64): === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 20.891 ns/iter 2 thread contention: 211.037 ns/iter 3 thread contention: 409.962 ns/iter 4 thread contention: 657.441 ns/iter 5 thread contention: 828.405 ns/iter 6 thread contention: 1131.827 ns/iter => std::mutex No contention: 50.884 ns/iter 2 thread contention: 103.620 ns/iter 3 thread contention: 332.429 ns/iter 4 thread contention: 620.802 ns/iter 5 thread contention: 783.943 ns/iter 6 thread contention: 834.002 ns/iter => std::shared_timed_mutex No contention: 64.948 ns/iter 2 thread contention: 173.191 ns/iter 3 thread contention: 490.352 ns/iter 4 thread contention: 660.668 ns/iter 5 thread contention: 1014.546 ns/iter 6 thread contention: 1451.553 ns/iter => std::shared_mutex No contention: 64.521 ns/iter 2 thread contention: 195.222 ns/iter 3 thread contention: 490.819 ns/iter 4 thread contention: 654.786 ns/iter 5 thread contention: 955.759 ns/iter 6 thread contention: 1282.544 ns/iter
2018-07-13 22:39:01 +00:00
TracyMutex lock;
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
std::vector<float> mbps;
float compRatio;
};
Handle ZoneBeginCallstack queue event. This is identical to ZoneBegin handling, but requires some additional bookkeeping to account for the incoming callstack information.
2018-06-21 23:07:25 +00:00
enum class NextCallstackType
{
Process GpuZoneBeginCallstack queue event.
2018-06-21 23:56:32 +00:00
Zone,
Send crash report.
2018-08-20 00:07:31 +00:00
Gpu,
Crash
Handle ZoneBeginCallstack queue event. This is identical to ZoneBegin handling, but requires some additional bookkeeping to account for the incoming callstack information.
2018-06-21 23:07:25 +00:00
};
struct NextCallstack
{
NextCallstackType type;
union
{
ZoneEvent* zone;
Process GpuZoneBeginCallstack queue event.
2018-06-21 23:56:32 +00:00
GpuEvent* gpu;
Handle ZoneBeginCallstack queue event. This is identical to ZoneBegin handling, but requires some additional bookkeeping to account for the incoming callstack information.
2018-06-21 23:07:25 +00:00
};
};
Gracefully store failure states.
2019-01-14 22:22:31 +00:00
struct FailureData
{
uint64_t thread;
int32_t srcloc;
};
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
public:
Gracefully store failure states.
2019-01-14 22:22:31 +00:00
enum class Failure
{
None,
Store failure reason strings in Worker.
2019-01-15 17:42:15 +00:00
ZoneStack,
Detect zone end without begin failure.
2019-01-15 23:45:48 +00:00
ZoneEnd,
Add ZoneText and ZoneName to the C API.
2019-01-16 00:17:01 +00:00
ZoneText,
ZoneName,
Track memory alloc+free matching failures.
2019-01-15 17:56:26 +00:00
MemFree,
Handle discontinuous frame end mismatch failure.
2019-02-28 18:32:42 +00:00
FrameEnd,
Store failure reason strings in Worker.
2019-01-15 17:42:15 +00:00
NUM_FAILURES
Gracefully store failure states.
2019-01-14 22:22:31 +00:00
};
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
Worker( const char* addr );
Allow partial load of data from dump.
2018-04-20 14:03:09 +00:00
Worker( FileRead& f, EventType::Type eventMask = EventType::All );
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
~Worker();
const std::string& GetAddr() const { return m_addr; }
const std::string& GetCaptureName() const { return m_captureName; }
Store explicit program name and capture time.
2018-08-28 22:57:11 +00:00
const std::string& GetCaptureProgram() const { return m_captureProgram; }
uint64_t GetCaptureTime() const { return m_captureTime; }
Display host info.
2018-08-19 16:24:43 +00:00
const std::string& GetHostInfo() const { return m_hostInfo; }
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
int64_t GetDelay() const { return m_delay; }
int64_t GetResolution() const { return m_resolution; }
Use the fastest mutex available. The selection is based on the following test results: MSVC: === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 11.641 ns/iter 2 thread contention: 141.559 ns/iter 3 thread contention: 242.733 ns/iter 4 thread contention: 409.807 ns/iter 5 thread contention: 561.544 ns/iter 6 thread contention: 785.845 ns/iter => std::mutex No contention: 19.190 ns/iter 2 thread contention: 39.305 ns/iter 3 thread contention: 58.999 ns/iter 4 thread contention: 59.532 ns/iter 5 thread contention: 103.539 ns/iter 6 thread contention: 110.314 ns/iter => std::shared_timed_mutex No contention: 45.487 ns/iter 2 thread contention: 96.351 ns/iter 3 thread contention: 142.871 ns/iter 4 thread contention: 184.999 ns/iter 5 thread contention: 336.608 ns/iter 6 thread contention: 542.551 ns/iter => std::shared_mutex No contention: 10.861 ns/iter 2 thread contention: 17.495 ns/iter 3 thread contention: 31.126 ns/iter 4 thread contention: 40.468 ns/iter 5 thread contention: 15.677 ns/iter 6 thread contention: 64.505 ns/iter Cygwin (clang): === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 11.536 ns/iter 2 thread contention: 121.082 ns/iter 3 thread contention: 396.430 ns/iter 4 thread contention: 672.555 ns/iter 5 thread contention: 1327.761 ns/iter 6 thread contention: 14151.955 ns/iter => std::mutex No contention: 62.583 ns/iter 2 thread contention: 3990.464 ns/iter 3 thread contention: 7161.189 ns/iter 4 thread contention: 9870.820 ns/iter 5 thread contention: 12355.178 ns/iter 6 thread contention: 14694.903 ns/iter => std::shared_timed_mutex No contention: 91.687 ns/iter 2 thread contention: 1115.037 ns/iter 3 thread contention: 4183.792 ns/iter 4 thread contention: 15283.491 ns/iter 5 thread contention: 27812.477 ns/iter 6 thread contention: 35028.140 ns/iter => std::shared_mutex No contention: 91.764 ns/iter 2 thread contention: 1051.826 ns/iter 3 thread contention: 5574.720 ns/iter 4 thread contention: 15721.416 ns/iter 5 thread contention: 27721.487 ns/iter 6 thread contention: 35420.404 ns/iter Linux (x64): === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 13.487 ns/iter 2 thread contention: 210.317 ns/iter 3 thread contention: 430.855 ns/iter 4 thread contention: 510.533 ns/iter 5 thread contention: 1003.609 ns/iter 6 thread contention: 1787.683 ns/iter => std::mutex No contention: 12.403 ns/iter 2 thread contention: 157.122 ns/iter 3 thread contention: 186.791 ns/iter 4 thread contention: 265.073 ns/iter 5 thread contention: 283.778 ns/iter 6 thread contention: 270.687 ns/iter => std::shared_timed_mutex No contention: 21.509 ns/iter 2 thread contention: 150.179 ns/iter 3 thread contention: 256.574 ns/iter 4 thread contention: 415.351 ns/iter 5 thread contention: 611.532 ns/iter 6 thread contention: 944.695 ns/iter => std::shared_mutex No contention: 20.805 ns/iter 2 thread contention: 157.034 ns/iter 3 thread contention: 244.025 ns/iter 4 thread contention: 406.269 ns/iter 5 thread contention: 387.985 ns/iter 6 thread contention: 468.550 ns/iter Linux (arm64): === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 20.891 ns/iter 2 thread contention: 211.037 ns/iter 3 thread contention: 409.962 ns/iter 4 thread contention: 657.441 ns/iter 5 thread contention: 828.405 ns/iter 6 thread contention: 1131.827 ns/iter => std::mutex No contention: 50.884 ns/iter 2 thread contention: 103.620 ns/iter 3 thread contention: 332.429 ns/iter 4 thread contention: 620.802 ns/iter 5 thread contention: 783.943 ns/iter 6 thread contention: 834.002 ns/iter => std::shared_timed_mutex No contention: 64.948 ns/iter 2 thread contention: 173.191 ns/iter 3 thread contention: 490.352 ns/iter 4 thread contention: 660.668 ns/iter 5 thread contention: 1014.546 ns/iter 6 thread contention: 1451.553 ns/iter => std::shared_mutex No contention: 64.521 ns/iter 2 thread contention: 195.222 ns/iter 3 thread contention: 490.819 ns/iter 4 thread contention: 654.786 ns/iter 5 thread contention: 955.759 ns/iter 6 thread contention: 1282.544 ns/iter
2018-07-13 22:39:01 +00:00
TracyMutex& GetDataLock() { return m_data.lock; }
Multiple frame sets support.
2018-08-04 17:47:09 +00:00
size_t GetFrameCount( const FrameData& fd ) const { return fd.frames.size(); }
Proper way to get full frame count.
2018-09-01 10:38:12 +00:00
size_t GetFullFrameCount( const FrameData& fd ) const;
Make getting start of time less cryptic.
2018-08-04 17:51:50 +00:00
int64_t GetTimeBegin() const { return GetFrameBegin( *m_data.framesBase, 0 ); }
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
int64_t GetLastTime() const { return m_data.lastTime; }
uint64_t GetZoneCount() const { return m_data.zonesCnt; }
Add lock, plot counts to worker.
2018-08-08 17:21:53 +00:00
uint64_t GetLockCount() const;
uint64_t GetPlotCount() const;
Display srcloc, callstack counts.
2018-08-14 14:36:25 +00:00
uint64_t GetSrcLocCount() const { return m_data.sourceLocationPayload.size() + m_data.sourceLocation.size(); }
uint64_t GetCallstackPayloadCount() const { return m_data.callstackPayload.size() - 1; }
uint64_t GetCallstackFrameCount() const { return m_data.callstackFrameMap.size(); }
Store on-demand frame offset.
2018-07-10 20:39:41 +00:00
uint64_t GetFrameOffset() const { return m_data.frameOffset; }
Multiple frame sets support.
2018-08-04 17:47:09 +00:00
const FrameData* GetFramesBase() const { return m_data.framesBase; }
Add FrameData vector accessor.
2018-08-04 18:58:49 +00:00
const Vector<FrameData*>& GetFrames() const { return m_data.frames.Data(); }
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
Multiple frame sets support.
2018-08-04 17:47:09 +00:00
int64_t GetFrameTime( const FrameData& fd, size_t idx ) const;
int64_t GetFrameBegin( const FrameData& fd, size_t idx ) const;
int64_t GetFrameEnd( const FrameData& fd, size_t idx ) const;
std::pair <int, int> GetFrameRange( const FrameData& fd, int64_t from, int64_t to );
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
const std::map<uint32_t, LockMap>& GetLockMap() const { return m_data.lockMap; }
const Vector<MessageData*>& GetMessages() const { return m_data.messages; }
const Vector<GpuCtxData*>& GetGpuData() const { return m_data.gpuData; }
Use StringDiscovery for plots.
2018-08-04 14:33:03 +00:00
const Vector<PlotData*>& GetPlots() const { return m_data.plots.Data(); }
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
const Vector<ThreadData*>& GetThreadData() const { return m_data.threads; }
Memory data accessor.
2018-04-01 18:25:09 +00:00
const MemData& GetMemData() const { return m_data.memory; }
Stack frame accessor.
2018-06-19 23:18:59 +00:00
Replace frame pointers with callstack frame ids.
2019-03-03 15:50:18 +00:00
const VarArray<CallstackFrameId>& GetCallstack( uint32_t idx ) const { return *m_data.callstackPayload[idx]; }
const CallstackFrameData* GetCallstackFrame( const CallstackFrameId& ptr ) const;
Pack/unpack frame pointer to callstack frame id.
2019-03-03 15:39:13 +00:00
uint64_t GetCanonicalPointer( const CallstackFrameId& id ) const;
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
Display crash information in info window.
2018-08-20 00:23:29 +00:00
const CrashEvent& GetCrashEvent() const { return m_data.m_crashEvent; }
Add direct zone end getter.
2018-03-18 01:53:00 +00:00
// Some zones may have incomplete timing data (only start time is available, end hasn't arrived yet).
// GetZoneEnd() will try to infer the end time by looking at child zones (parent zone can't end
// before its children have ended).
// GetZoneEndDirect() will only return zone's direct timing data, without looking at children.
Store children vectors in a separate data collection. This reduces per-zone memory cost by 9 bytes if there are no children and increases it by 4 bytes, if there are children. This is universally a better solution, as the following data shows: +++ /home/wolf/desktop/tracy-old/android.tracy +++ Vectors: 2794480 Size 0: 2373070 (84.92%) Size 1: 70237 (2.51%) Size 2+: 351173 (12.57%) +++ /home/wolf/desktop/tracy-old/asset-new.tracy +++ Vectors: 1799227 Size 0: 1482691 (82.41%) Size 1: 93272 (5.18%) Size 2+: 223264 (12.41%) +++ /home/wolf/desktop/tracy-old/asset-new-id.tracy +++ Vectors: 1977996 Size 0: 1640817 (82.95%) Size 1: 97198 (4.91%) Size 2+: 239981 (12.13%) +++ /home/wolf/desktop/tracy-old/asset-old.tracy +++ Vectors: 1782395 Size 0: 1471437 (82.55%) Size 1: 88813 (4.98%) Size 2+: 222145 (12.46%) +++ /home/wolf/desktop/tracy-old/big.tracy +++ Vectors: 180794047 Size 0: 172696094 (95.52%) Size 1: 2799772 (1.55%) Size 2+: 5298181 (2.93%) +++ /home/wolf/desktop/tracy-old/darkrl.tracy +++ Vectors: 12014129 Size 0: 11611324 (96.65%) Size 1: 134980 (1.12%) Size 2+: 267825 (2.23%) +++ /home/wolf/desktop/tracy-old/mem.tracy +++ Vectors: 383097 Size 0: 321932 (84.03%) Size 1: 854 (0.22%) Size 2+: 60311 (15.74%) +++ /home/wolf/desktop/tracy-old/new.tracy +++ Vectors: 77536 Size 0: 63035 (81.30%) Size 1: 8886 (11.46%) Size 2+: 5615 (7.24%) +++ /home/wolf/desktop/tracy-old/selfprofile.tracy +++ Vectors: 22940871 Size 0: 22704868 (98.97%) Size 1: 73000 (0.32%) Size 2+: 163003 (0.71%) +++ /home/wolf/desktop/tracy-old/tbrowser.tracy +++ Vectors: 962682 Size 0: 695380 (72.23%) Size 1: 43007 (4.47%) Size 2+: 224295 (23.30%) +++ /home/wolf/desktop/tracy-old/virtualfile_hc.tracy +++ Vectors: 529170 Size 0: 449386 (84.92%) Size 1: 15694 (2.97%) Size 2+: 64090 (12.11%) +++ /home/wolf/desktop/tracy-old/zfile_hc.tracy +++ Vectors: 264849 Size 0: 220589 (83.29%) Size 1: 9386 (3.54%) Size 2+: 34874 (13.17%)
2018-07-22 14:05:50 +00:00
int64_t GetZoneEnd( const ZoneEvent& ev );
Also remove child vectors from GPU events.
2018-07-22 17:47:01 +00:00
int64_t GetZoneEnd( const GpuEvent& ev );
Change zone end "-1" comparisons to "0" comparisons.
2018-03-24 01:00:20 +00:00
static tracy_force_inline int64_t GetZoneEndDirect( const ZoneEvent& ev ) { return ev.end >= 0 ? ev.end : ev.start; }
static tracy_force_inline int64_t GetZoneEndDirect( const GpuEvent& ev ) { return ev.gpuEnd >= 0 ? ev.gpuEnd : ev.gpuStart; }
Add direct zone end getter.
2018-03-18 01:53:00 +00:00
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
const char* GetString( uint64_t ptr ) const;
const char* GetString( const StringRef& ref ) const;
const char* GetString( const StringIdx& idx ) const;
const char* GetThreadString( uint64_t id ) const;
const SourceLocation& GetSourceLocation( int32_t srcloc ) const;
Allow zone name retrieval from source location.
2019-02-10 15:45:19 +00:00
const char* GetZoneName( const SourceLocation& srcloc ) const;
Extract zone name getting functionality.
2018-06-29 13:14:20 +00:00
const char* GetZoneName( const ZoneEvent& ev ) const;
const char* GetZoneName( const ZoneEvent& ev, const SourceLocation& srcloc ) const;
const char* GetZoneName( const GpuEvent& ev ) const;
const char* GetZoneName( const GpuEvent& ev, const SourceLocation& srcloc ) const;
Store children vectors in a separate data collection. This reduces per-zone memory cost by 9 bytes if there are no children and increases it by 4 bytes, if there are children. This is universally a better solution, as the following data shows: +++ /home/wolf/desktop/tracy-old/android.tracy +++ Vectors: 2794480 Size 0: 2373070 (84.92%) Size 1: 70237 (2.51%) Size 2+: 351173 (12.57%) +++ /home/wolf/desktop/tracy-old/asset-new.tracy +++ Vectors: 1799227 Size 0: 1482691 (82.41%) Size 1: 93272 (5.18%) Size 2+: 223264 (12.41%) +++ /home/wolf/desktop/tracy-old/asset-new-id.tracy +++ Vectors: 1977996 Size 0: 1640817 (82.95%) Size 1: 97198 (4.91%) Size 2+: 239981 (12.13%) +++ /home/wolf/desktop/tracy-old/asset-old.tracy +++ Vectors: 1782395 Size 0: 1471437 (82.55%) Size 1: 88813 (4.98%) Size 2+: 222145 (12.46%) +++ /home/wolf/desktop/tracy-old/big.tracy +++ Vectors: 180794047 Size 0: 172696094 (95.52%) Size 1: 2799772 (1.55%) Size 2+: 5298181 (2.93%) +++ /home/wolf/desktop/tracy-old/darkrl.tracy +++ Vectors: 12014129 Size 0: 11611324 (96.65%) Size 1: 134980 (1.12%) Size 2+: 267825 (2.23%) +++ /home/wolf/desktop/tracy-old/mem.tracy +++ Vectors: 383097 Size 0: 321932 (84.03%) Size 1: 854 (0.22%) Size 2+: 60311 (15.74%) +++ /home/wolf/desktop/tracy-old/new.tracy +++ Vectors: 77536 Size 0: 63035 (81.30%) Size 1: 8886 (11.46%) Size 2+: 5615 (7.24%) +++ /home/wolf/desktop/tracy-old/selfprofile.tracy +++ Vectors: 22940871 Size 0: 22704868 (98.97%) Size 1: 73000 (0.32%) Size 2+: 163003 (0.71%) +++ /home/wolf/desktop/tracy-old/tbrowser.tracy +++ Vectors: 962682 Size 0: 695380 (72.23%) Size 1: 43007 (4.47%) Size 2+: 224295 (23.30%) +++ /home/wolf/desktop/tracy-old/virtualfile_hc.tracy +++ Vectors: 529170 Size 0: 449386 (84.92%) Size 1: 15694 (2.97%) Size 2+: 64090 (12.11%) +++ /home/wolf/desktop/tracy-old/zfile_hc.tracy +++ Vectors: 264849 Size 0: 220589 (83.29%) Size 1: 9386 (3.54%) Size 2+: 34874 (13.17%)
2018-07-22 14:05:50 +00:00
tracy_force_inline const Vector<ZoneEvent*>& GetZoneChildren( int32_t idx ) const { return m_data.m_zoneChildren[idx]; }
Also remove child vectors from GPU events.
2018-07-22 17:47:01 +00:00
tracy_force_inline const Vector<GpuEvent*>& GetGpuChildren( int32_t idx ) const { return m_data.m_gpuChildren[idx]; }
Store children vectors in a separate data collection. This reduces per-zone memory cost by 9 bytes if there are no children and increases it by 4 bytes, if there are children. This is universally a better solution, as the following data shows: +++ /home/wolf/desktop/tracy-old/android.tracy +++ Vectors: 2794480 Size 0: 2373070 (84.92%) Size 1: 70237 (2.51%) Size 2+: 351173 (12.57%) +++ /home/wolf/desktop/tracy-old/asset-new.tracy +++ Vectors: 1799227 Size 0: 1482691 (82.41%) Size 1: 93272 (5.18%) Size 2+: 223264 (12.41%) +++ /home/wolf/desktop/tracy-old/asset-new-id.tracy +++ Vectors: 1977996 Size 0: 1640817 (82.95%) Size 1: 97198 (4.91%) Size 2+: 239981 (12.13%) +++ /home/wolf/desktop/tracy-old/asset-old.tracy +++ Vectors: 1782395 Size 0: 1471437 (82.55%) Size 1: 88813 (4.98%) Size 2+: 222145 (12.46%) +++ /home/wolf/desktop/tracy-old/big.tracy +++ Vectors: 180794047 Size 0: 172696094 (95.52%) Size 1: 2799772 (1.55%) Size 2+: 5298181 (2.93%) +++ /home/wolf/desktop/tracy-old/darkrl.tracy +++ Vectors: 12014129 Size 0: 11611324 (96.65%) Size 1: 134980 (1.12%) Size 2+: 267825 (2.23%) +++ /home/wolf/desktop/tracy-old/mem.tracy +++ Vectors: 383097 Size 0: 321932 (84.03%) Size 1: 854 (0.22%) Size 2+: 60311 (15.74%) +++ /home/wolf/desktop/tracy-old/new.tracy +++ Vectors: 77536 Size 0: 63035 (81.30%) Size 1: 8886 (11.46%) Size 2+: 5615 (7.24%) +++ /home/wolf/desktop/tracy-old/selfprofile.tracy +++ Vectors: 22940871 Size 0: 22704868 (98.97%) Size 1: 73000 (0.32%) Size 2+: 163003 (0.71%) +++ /home/wolf/desktop/tracy-old/tbrowser.tracy +++ Vectors: 962682 Size 0: 695380 (72.23%) Size 1: 43007 (4.47%) Size 2+: 224295 (23.30%) +++ /home/wolf/desktop/tracy-old/virtualfile_hc.tracy +++ Vectors: 529170 Size 0: 449386 (84.92%) Size 1: 15694 (2.97%) Size 2+: 64090 (12.11%) +++ /home/wolf/desktop/tracy-old/zfile_hc.tracy +++ Vectors: 264849 Size 0: 220589 (83.29%) Size 1: 9386 (3.54%) Size 2+: 34874 (13.17%)
2018-07-22 14:05:50 +00:00
Add support for matching source locations ignoring case.
2018-12-18 15:52:29 +00:00
std::vector<int32_t> GetMatchingSourceLocation( const char* query, bool ignoreCase ) const;
Allow raw access to source location zones data.
2018-03-24 13:48:52 +00:00
Reduce data collection if TRACY_NO_STATISTICS is defined. Statistical data collection is only useful if it's meant to be used. Otherwise it only incurs CPU and memory cost.
2018-03-18 11:55:54 +00:00
#ifndef TRACY_NO_STATISTICS
Use precalculated min/max time spans.
2018-03-18 19:20:24 +00:00
const SourceLocationZones& GetZonesForSourceLocation( int32_t srcloc ) const;
Allow raw access to source location zones data.
2018-03-24 13:48:52 +00:00
const flat_hash_map<int32_t, SourceLocationZones, nohash<int32_t>>& GetSourceLocationZones() const { return m_data.sourceLocationZones; }
Sort source location zones on a separate thread.
2018-04-30 01:54:09 +00:00
bool AreSourceLocationZonesReady() const { return m_data.sourceLocationZonesReady; }
Reduce data collection if TRACY_NO_STATISTICS is defined. Statistical data collection is only useful if it's meant to be used. Otherwise it only incurs CPU and memory cost.
2018-03-18 11:55:54 +00:00
#endif
Implement search for matching source locations.
2018-03-04 15:52:45 +00:00
CompressThread body must be available.
2018-05-03 16:43:51 +00:00
tracy_force_inline uint16_t CompressThread( uint64_t thread )
{
if( m_data.threadLast.first == thread ) return m_data.threadLast.second;
return CompressThreadReal( thread );
}
Force inline compress/decompress thread id.
2018-03-24 00:31:58 +00:00
tracy_force_inline uint64_t DecompressThread( uint16_t thread ) const { assert( thread < m_data.threadExpand.size() ); return m_data.threadExpand[thread]; }
Thread id compression/decompression.
2018-03-18 19:45:22 +00:00
Use the fastest mutex available. The selection is based on the following test results: MSVC: === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 11.641 ns/iter 2 thread contention: 141.559 ns/iter 3 thread contention: 242.733 ns/iter 4 thread contention: 409.807 ns/iter 5 thread contention: 561.544 ns/iter 6 thread contention: 785.845 ns/iter => std::mutex No contention: 19.190 ns/iter 2 thread contention: 39.305 ns/iter 3 thread contention: 58.999 ns/iter 4 thread contention: 59.532 ns/iter 5 thread contention: 103.539 ns/iter 6 thread contention: 110.314 ns/iter => std::shared_timed_mutex No contention: 45.487 ns/iter 2 thread contention: 96.351 ns/iter 3 thread contention: 142.871 ns/iter 4 thread contention: 184.999 ns/iter 5 thread contention: 336.608 ns/iter 6 thread contention: 542.551 ns/iter => std::shared_mutex No contention: 10.861 ns/iter 2 thread contention: 17.495 ns/iter 3 thread contention: 31.126 ns/iter 4 thread contention: 40.468 ns/iter 5 thread contention: 15.677 ns/iter 6 thread contention: 64.505 ns/iter Cygwin (clang): === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 11.536 ns/iter 2 thread contention: 121.082 ns/iter 3 thread contention: 396.430 ns/iter 4 thread contention: 672.555 ns/iter 5 thread contention: 1327.761 ns/iter 6 thread contention: 14151.955 ns/iter => std::mutex No contention: 62.583 ns/iter 2 thread contention: 3990.464 ns/iter 3 thread contention: 7161.189 ns/iter 4 thread contention: 9870.820 ns/iter 5 thread contention: 12355.178 ns/iter 6 thread contention: 14694.903 ns/iter => std::shared_timed_mutex No contention: 91.687 ns/iter 2 thread contention: 1115.037 ns/iter 3 thread contention: 4183.792 ns/iter 4 thread contention: 15283.491 ns/iter 5 thread contention: 27812.477 ns/iter 6 thread contention: 35028.140 ns/iter => std::shared_mutex No contention: 91.764 ns/iter 2 thread contention: 1051.826 ns/iter 3 thread contention: 5574.720 ns/iter 4 thread contention: 15721.416 ns/iter 5 thread contention: 27721.487 ns/iter 6 thread contention: 35420.404 ns/iter Linux (x64): === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 13.487 ns/iter 2 thread contention: 210.317 ns/iter 3 thread contention: 430.855 ns/iter 4 thread contention: 510.533 ns/iter 5 thread contention: 1003.609 ns/iter 6 thread contention: 1787.683 ns/iter => std::mutex No contention: 12.403 ns/iter 2 thread contention: 157.122 ns/iter 3 thread contention: 186.791 ns/iter 4 thread contention: 265.073 ns/iter 5 thread contention: 283.778 ns/iter 6 thread contention: 270.687 ns/iter => std::shared_timed_mutex No contention: 21.509 ns/iter 2 thread contention: 150.179 ns/iter 3 thread contention: 256.574 ns/iter 4 thread contention: 415.351 ns/iter 5 thread contention: 611.532 ns/iter 6 thread contention: 944.695 ns/iter => std::shared_mutex No contention: 20.805 ns/iter 2 thread contention: 157.034 ns/iter 3 thread contention: 244.025 ns/iter 4 thread contention: 406.269 ns/iter 5 thread contention: 387.985 ns/iter 6 thread contention: 468.550 ns/iter Linux (arm64): === Lock test, 6 threads === => NonRecursiveBenaphore No contention: 20.891 ns/iter 2 thread contention: 211.037 ns/iter 3 thread contention: 409.962 ns/iter 4 thread contention: 657.441 ns/iter 5 thread contention: 828.405 ns/iter 6 thread contention: 1131.827 ns/iter => std::mutex No contention: 50.884 ns/iter 2 thread contention: 103.620 ns/iter 3 thread contention: 332.429 ns/iter 4 thread contention: 620.802 ns/iter 5 thread contention: 783.943 ns/iter 6 thread contention: 834.002 ns/iter => std::shared_timed_mutex No contention: 64.948 ns/iter 2 thread contention: 173.191 ns/iter 3 thread contention: 490.352 ns/iter 4 thread contention: 660.668 ns/iter 5 thread contention: 1014.546 ns/iter 6 thread contention: 1451.553 ns/iter => std::shared_mutex No contention: 64.521 ns/iter 2 thread contention: 195.222 ns/iter 3 thread contention: 490.819 ns/iter 4 thread contention: 654.786 ns/iter 5 thread contention: 955.759 ns/iter 6 thread contention: 1282.544 ns/iter
2018-07-13 22:39:01 +00:00
TracyMutex& GetMbpsDataLock() { return m_mbpsData.lock; }
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
const std::vector<float>& GetMbpsData() const { return m_mbpsData.mbps; }
float GetCompRatio() const { return m_mbpsData.compRatio; }
bool HasData() const { return m_hasData.load( std::memory_order_acquire ); }
bool IsConnected() const { return m_connected.load( std::memory_order_relaxed ); }
Expose worker data state (static/dynamic).
2018-03-24 13:43:57 +00:00
bool IsDataStatic() const { return !m_thread.joinable(); }
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
void Shutdown() { m_shutdown.store( true, std::memory_order_relaxed ); }
Add placeholder worker disconnect command.
2019-01-24 17:51:55 +00:00
void Disconnect() { Shutdown(); } // TODO: Needs proper implementation.
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
void Write( FileWrite& f );
Track trace version in worker.
2018-07-29 13:33:48 +00:00
int GetTraceVersion() const { return m_traceVersion; }
Send protocol version to verify handshake.
2018-09-09 17:28:53 +00:00
uint8_t GetHandshakeStatus() const { return m_handshake.load( std::memory_order_relaxed ); }
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
Track trace loading progress.
2018-07-28 15:59:17 +00:00
static const LoadProgress& GetLoadProgress() { return s_loadProgress; }
Track trace loading time.
2019-01-06 18:09:50 +00:00
int64_t GetLoadTime() const { return m_loadTime; }
Track trace loading progress.
2018-07-28 15:59:17 +00:00
Gracefully store failure states.
2019-01-14 22:22:31 +00:00
void ClearFailure() { m_failure = Failure::None; }
Failure GetFailureType() const { return m_failure; }
const FailureData& GetFailureData() const { return m_failureData; }
Store failure reason strings in Worker.
2019-01-15 17:42:15 +00:00
static const char* GetFailureString( Failure failure );
Gracefully store failure states.
2019-01-14 22:22:31 +00:00
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
private:
void Exec();
void ServerQuery( uint8_t type, uint64_t data );
Perform zone stack validation.
2019-01-14 22:08:34 +00:00
tracy_force_inline bool DispatchProcess( const QueueItem& ev, char*& ptr );
tracy_force_inline bool Process( const QueueItem& ev );
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
tracy_force_inline void ProcessZoneBegin( const QueueZoneBegin& ev );
Handle ZoneBeginCallstack queue event. This is identical to ZoneBegin handling, but requires some additional bookkeeping to account for the incoming callstack information.
2018-06-21 23:07:25 +00:00
tracy_force_inline void ProcessZoneBeginCallstack( const QueueZoneBegin& ev );
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
tracy_force_inline void ProcessZoneBeginAllocSrcLoc( const QueueZoneBegin& ev );
Separate message for zone begin with alloc src loc and callstack.
2019-03-03 16:47:26 +00:00
tracy_force_inline void ProcessZoneBeginAllocSrcLocCallstack( const QueueZoneBegin& ev );
Simplify failure detection code.
2019-01-15 17:55:21 +00:00
tracy_force_inline void ProcessZoneEnd( const QueueZoneEnd& ev );
Process zone validation messages.
2019-01-14 21:56:10 +00:00
tracy_force_inline void ProcessZoneValidation( const QueueZoneValidation& ev );
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
tracy_force_inline void ProcessFrameMark( const QueueFrameMark& ev );
Add support for discontinuous frames.
2018-08-05 00:09:59 +00:00
tracy_force_inline void ProcessFrameMarkStart( const QueueFrameMark& ev );
tracy_force_inline void ProcessFrameMarkEnd( const QueueFrameMark& ev );
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
tracy_force_inline void ProcessZoneText( const QueueZoneText& ev );
Process custom free-form zone names.
2018-06-29 14:12:17 +00:00
tracy_force_inline void ProcessZoneName( const QueueZoneText& ev );
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
tracy_force_inline void ProcessLockAnnounce( const QueueLockAnnounce& ev );
No-op processing of lock terminate events.
2018-12-16 19:39:30 +00:00
tracy_force_inline void ProcessLockTerminate( const QueueLockTerminate& ev );
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
tracy_force_inline void ProcessLockWait( const QueueLockWait& ev );
tracy_force_inline void ProcessLockObtain( const QueueLockObtain& ev );
tracy_force_inline void ProcessLockRelease( const QueueLockRelease& ev );
tracy_force_inline void ProcessLockSharedWait( const QueueLockWait& ev );
tracy_force_inline void ProcessLockSharedObtain( const QueueLockObtain& ev );
tracy_force_inline void ProcessLockSharedRelease( const QueueLockRelease& ev );
tracy_force_inline void ProcessLockMark( const QueueLockMark& ev );
tracy_force_inline void ProcessPlotData( const QueuePlotData& ev );
tracy_force_inline void ProcessMessage( const QueueMessage& ev );
tracy_force_inline void ProcessMessageLiteral( const QueueMessage& ev );
tracy_force_inline void ProcessGpuNewContext( const QueueGpuNewContext& ev );
tracy_force_inline void ProcessGpuZoneBegin( const QueueGpuZoneBegin& ev );
Process GpuZoneBeginCallstack queue event.
2018-06-21 23:56:32 +00:00
tracy_force_inline void ProcessGpuZoneBeginCallstack( const QueueGpuZoneBegin& ev );
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
tracy_force_inline void ProcessGpuZoneEnd( const QueueGpuZoneEnd& ev );
tracy_force_inline void ProcessGpuTime( const QueueGpuTime& ev );
Broken memory events processing.
2018-04-01 00:03:34 +00:00
tracy_force_inline void ProcessMemAlloc( const QueueMemAlloc& ev );
Ignore dangling memory frees in on-demand mode.
2018-07-11 23:35:32 +00:00
tracy_force_inline bool ProcessMemFree( const QueueMemFree& ev );
Process memory events with callstack.
2018-06-19 16:52:45 +00:00
tracy_force_inline void ProcessMemAllocCallstack( const QueueMemAlloc& ev );
tracy_force_inline void ProcessMemFreeCallstack( const QueueMemFree& ev );
tracy_force_inline void ProcessCallstackMemory( const QueueCallstackMemory& ev );
Process callstack queue event.
2018-06-21 23:15:49 +00:00
tracy_force_inline void ProcessCallstack( const QueueCallstack& ev );
Separate processing for allocated callstacks. Only native callstack is used for the moment.
2019-03-05 01:04:45 +00:00
tracy_force_inline void ProcessCallstackAlloc( const QueueCallstackAlloc& ev );
Make it possible to store multiple frames at single frame address.
2019-01-20 18:11:48 +00:00
tracy_force_inline void ProcessCallstackFrameSize( const QueueCallstackFrameSize& ev );
Process callstack frames.
2018-06-19 23:07:09 +00:00
tracy_force_inline void ProcessCallstackFrame( const QueueCallstackFrame& ev );
Send crash report.
2018-08-20 00:07:31 +00:00
tracy_force_inline void ProcessCrashReport( const QueueCrashReport& ev );
Process CPU usage reports.
2019-02-21 21:45:39 +00:00
tracy_force_inline void ProcessSysTime( const QueueSysTime& ev );
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
Handle ZoneBeginCallstack queue event. This is identical to ZoneBegin handling, but requires some additional bookkeeping to account for the incoming callstack information.
2018-06-21 23:07:25 +00:00
tracy_force_inline void ProcessZoneBeginImpl( ZoneEvent* zone, const QueueZoneBegin& ev );
Separate message for zone begin with alloc src loc and callstack.
2019-03-03 16:47:26 +00:00
tracy_force_inline void ProcessZoneBeginAllocSrcLocImpl( ZoneEvent* zone, const QueueZoneBegin& ev );
Process GpuZoneBeginCallstack queue event.
2018-06-21 23:56:32 +00:00
tracy_force_inline void ProcessGpuZoneBeginImpl( GpuEvent* zone, const QueueGpuZoneBegin& ev );
Handle ZoneBeginCallstack queue event. This is identical to ZoneBegin handling, but requires some additional bookkeeping to account for the incoming callstack information.
2018-06-21 23:07:25 +00:00
Gracefully store failure states.
2019-01-14 22:22:31 +00:00
void ZoneStackFailure( uint64_t thread, const ZoneEvent* ev );
Detect zone end without begin failure.
2019-01-15 23:45:48 +00:00
void ZoneEndFailure( uint64_t thread );
Add ZoneText and ZoneName to the C API.
2019-01-16 00:17:01 +00:00
void ZoneTextFailure( uint64_t thread );
void ZoneNameFailure( uint64_t thread );
Track memory alloc+free matching failures.
2019-01-15 17:56:26 +00:00
void MemFreeFailure( uint64_t thread );
Handle discontinuous frame end mismatch failure.
2019-02-28 18:32:42 +00:00
void FrameEndFailure();
Gracefully store failure states.
2019-01-14 22:22:31 +00:00
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
tracy_force_inline void CheckSourceLocation( uint64_t ptr );
void NewSourceLocation( uint64_t ptr );
tracy_force_inline uint32_t ShrinkSourceLocation( uint64_t srcloc );
uint32_t NewShrinkedSourceLocation( uint64_t srcloc );
Create memory plot based on memory alloc/free events.
2018-04-28 13:49:12 +00:00
tracy_force_inline void MemAllocChanged( int64_t time );
void CreateMemAllocPlot();
Move memory plot reconstruction to a background thread.
2018-04-29 11:40:04 +00:00
void ReconstructMemAllocPlot();
Create memory plot based on memory alloc/free events.
2018-04-28 13:49:12 +00:00
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
void InsertMessageData( MessageData* msg, uint64_t thread );
ThreadData* NewThread( uint64_t thread );
ThreadData* NoticeThread( uint64_t thread );
tracy_force_inline void NewZone( ZoneEvent* zone, uint64_t thread );
void InsertLockEvent( LockMap& lockmap, LockEvent* lev, uint64_t thread );
void CheckString( uint64_t ptr );
void CheckThreadString( uint64_t id );
void AddSourceLocation( const QueueSourceLocation& srcloc );
void AddSourceLocationPayload( uint64_t ptr, char* data, size_t sz );
void AddString( uint64_t ptr, char* str, size_t sz );
void AddThreadString( uint64_t id, char* str, size_t sz );
void AddCustomString( uint64_t ptr, char* str, size_t sz );
Force inline AddCallstackPayload.
2018-06-24 13:28:09 +00:00
tracy_force_inline void AddCallstackPayload( uint64_t ptr, char* data, size_t sz );
Store callstack alloc payloads.
2019-03-03 16:34:56 +00:00
tracy_force_inline void AddCallstackAllocPayload( uint64_t ptr, char* data, size_t sz );
Store unique callstack payloads.
2018-06-19 19:15:36 +00:00
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
void InsertPlot( PlotData* plot, int64_t time, double val );
void HandlePlotName( uint64_t name, char* str, size_t sz );
Handle frame name queries.
2018-08-04 18:48:21 +00:00
void HandleFrameName( uint64_t name, char* str, size_t sz );
Broken memory events processing.
2018-04-01 00:03:34 +00:00
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
void HandlePostponedPlots();
StringLocation StoreString( char* str, size_t sz );
Add thread compression cache. Observation: calls to CompressThread() are likely to be repeated with the same value. Exploit that by storing last query and its result.
2018-04-30 23:29:25 +00:00
uint16_t CompressThreadReal( uint64_t thread );
Force inline compress/decompress thread id.
2018-03-24 00:31:58 +00:00
uint16_t CompressThreadNew( uint64_t thread );
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
Store time deltas, instead of absolute time in trace dumps. This change greatly reduces the size of saved dumps, but increase the cost of processing during loading. One notable outlier in the dataset below is mem.tracy, which increased in size, even if changes in the memory dump saving scheme decrease size of the other traces. 041/aa.tracy (0.4.1) {18987 KB} -> 042/aa.tracy (0.4.2) {10140 KB} 53.40% size change 041/android.tracy (0.4.1) {696753 KB} -> 042/android.tracy (0.4.2) {542738 KB} 77.90% size change 041/asset-new.tracy (0.4.1) {97163 KB} -> 042/asset-new.tracy (0.4.2) {78402 KB} 80.69% size change 041/asset-new-id.tracy (0.4.1) {105683 KB} -> 042/asset-new-id.tracy (0.4.2) {84341 KB} 79.81% size change 041/asset-old.tracy (0.4.1) {100205 KB} -> 042/asset-old.tracy (0.4.2) {80688 KB} 80.52% size change 041/big.tracy (0.4.1) {2246014 KB} -> 042/big.tracy (0.4.2) {943083 KB} 41.99% size change 041/crash.tracy (0.4.1) {143 KB} -> 042/crash.tracy (0.4.2) {131 KB} 91.39% size change 041/crash2.tracy (0.4.1) {3411 KB} -> 042/crash2.tracy (0.4.2) {1425 KB} 41.80% size change 041/darkrl.tracy (0.4.1) {31818 KB} -> 042/darkrl.tracy (0.4.2) {15897 KB} 49.96% size change 041/darkrl2.tracy (0.4.1) {18778 KB} -> 042/darkrl2.tracy (0.4.2) {8002 KB} 42.62% size change 041/darkrl-old.tracy (0.4.1) {151346 KB} -> 042/darkrl-old.tracy (0.4.2) {67945 KB} 44.89% size change 041/deadlock.tracy (0.4.1) {53 KB} -> 042/deadlock.tracy (0.4.2) {52 KB} 98.55% size change 041/gn-opengl.tracy (0.4.1) {45860 KB} -> 042/gn-opengl.tracy (0.4.2) {30983 KB} 67.56% size change 041/gn-vulkan.tracy (0.4.1) {45618 KB} -> 042/gn-vulkan.tracy (0.4.2) {31349 KB} 68.72% size change 041/long.tracy (0.4.1) {1583550 KB} -> 042/long.tracy (0.4.2) {1225316 KB} 77.38% size change 041/mem.tracy (0.4.1) {1243058 KB} -> 042/mem.tracy (0.4.2) {1369291 KB} 110.15% size change 041/multi.tracy (0.4.1) {14519 KB} -> 042/multi.tracy (0.4.2) {8110 KB} 55.86% size change 041/new.tracy (0.4.1) {1439 KB} -> 042/new.tracy (0.4.2) {1108 KB} 77.01% size change 041/q3bsp-mt.tracy (0.4.1) {1414323 KB} -> 042/q3bsp-mt.tracy (0.4.2) {949855 KB} 67.16% size change 041/q3bsp-st.tracy (0.4.1) {301334 KB} -> 042/q3bsp-st.tracy (0.4.2) {240347 KB} 79.76% size change 041/selfprofile.tracy (0.4.1) {399648 KB} -> 042/selfprofile.tracy (0.4.2) {197713 KB} 49.47% size change 041/tbrowser.tracy (0.4.1) {13052 KB} -> 042/tbrowser.tracy (0.4.2) {9503 KB} 72.81% size change 041/test.tracy (0.4.1) {60309 KB} -> 042/test.tracy (0.4.2) {40700 KB} 67.49% size change 041/virtualfile_hc.tracy (0.4.1) {108967 KB} -> 042/virtualfile_hc.tracy (0.4.2) {72839 KB} 66.85% size change 041/zfile_hc.tracy (0.4.1) {58814 KB} -> 042/zfile_hc.tracy (0.4.2) {39608 KB} 67.35% size change
2018-12-30 22:06:03 +00:00
tracy_force_inline void ReadTimeline( FileRead& f, ZoneEvent* zone, uint16_t thread, int64_t& refTime );
tracy_force_inline void ReadTimelinePre042( FileRead& f, ZoneEvent* zone, uint16_t thread, int fileVer );
Track separate time offset for GPU times. This is second version of 0.4.2 dump file format. Previous 0.4.2 format cannot be read anymore. 041/aa.tracy (0.4.1) {18987 KB} -> 042/aa.tracy (0.4.2) {10051 KB} 52.94% size change 041/android.tracy (0.4.1) {696753 KB} -> 042/android.tracy (0.4.2) {542738 KB} 77.90% size change 041/asset-new.tracy (0.4.1) {97163 KB} -> 042/asset-new.tracy (0.4.2) {78402 KB} 80.69% size change 041/asset-new-id.tracy (0.4.1) {105683 KB} -> 042/asset-new-id.tracy (0.4.2) {84341 KB} 79.81% size change 041/asset-old.tracy (0.4.1) {100205 KB} -> 042/asset-old.tracy (0.4.2) {80688 KB} 80.52% size change 041/big.tracy (0.4.1) {2246014 KB} -> 042/big.tracy (0.4.2) {939578 KB} 41.83% size change 041/crash.tracy (0.4.1) {143 KB} -> 042/crash.tracy (0.4.2) {131 KB} 91.37% size change 041/crash2.tracy (0.4.1) {3411 KB} -> 042/crash2.tracy (0.4.2) {1420 KB} 41.63% size change 041/darkrl.tracy (0.4.1) {31818 KB} -> 042/darkrl.tracy (0.4.2) {15762 KB} 49.54% size change 041/darkrl2.tracy (0.4.1) {18778 KB} -> 042/darkrl2.tracy (0.4.2) {7945 KB} 42.31% size change 041/darkrl-old.tracy (0.4.1) {151346 KB} -> 042/darkrl-old.tracy (0.4.2) {67449 KB} 44.57% size change 041/deadlock.tracy (0.4.1) {53 KB} -> 042/deadlock.tracy (0.4.2) {52 KB} 98.55% size change 041/gn-opengl.tracy (0.4.1) {45860 KB} -> 042/gn-opengl.tracy (0.4.2) {29005 KB} 63.25% size change 041/gn-vulkan.tracy (0.4.1) {45618 KB} -> 042/gn-vulkan.tracy (0.4.2) {29352 KB} 64.34% size change 041/long.tracy (0.4.1) {1583550 KB} -> 042/long.tracy (0.4.2) {1182800 KB} 74.69% size change 041/mem.tracy (0.4.1) {1243058 KB} -> 042/mem.tracy (0.4.2) {1369067 KB} 110.14% size change 041/multi.tracy (0.4.1) {14519 KB} -> 042/multi.tracy (0.4.2) {8000 KB} 55.10% size change 041/new.tracy (0.4.1) {1439 KB} -> 042/new.tracy (0.4.2) {1105 KB} 76.75% size change 041/q3bsp-mt.tracy (0.4.1) {1414323 KB} -> 042/q3bsp-mt.tracy (0.4.2) {949855 KB} 67.16% size change 041/q3bsp-st.tracy (0.4.1) {301334 KB} -> 042/q3bsp-st.tracy (0.4.2) {240347 KB} 79.76% size change 041/selfprofile.tracy (0.4.1) {399648 KB} -> 042/selfprofile.tracy (0.4.2) {197704 KB} 49.47% size change 041/tbrowser.tracy (0.4.1) {13052 KB} -> 042/tbrowser.tracy (0.4.2) {9503 KB} 72.81% size change 041/test.tracy (0.4.1) {60309 KB} -> 042/test.tracy (0.4.2) {40700 KB} 67.49% size change 041/virtualfile_hc.tracy (0.4.1) {108967 KB} -> 042/virtualfile_hc.tracy (0.4.2) {72424 KB} 66.46% size change 041/zfile_hc.tracy (0.4.1) {58814 KB} -> 042/zfile_hc.tracy (0.4.2) {39418 KB} 67.02% size change
2019-01-03 19:09:45 +00:00
tracy_force_inline void ReadTimeline( FileRead& f, GpuEvent* zone, int64_t& refTime, int64_t& refGpuTime );
Don't decompress GpuZone threads while saving trace. Saving the threads compressed in GPU zones and memory event has the following result on trace dump sizes: 043/aa.tracy (0.4.3) {10055 KB} -> 044/aa.tracy (0.4.4) {9975 KB} 99.20% size change 043/android.tracy (0.4.3) {542739 KB} -> 044/android.tracy (0.4.4) {519248 KB} 95.67% size change 043/asset-new.tracy (0.4.3) {78403 KB} -> 044/asset-new.tracy (0.4.4) {75899 KB} 96.81% size change 043/asset-new-id.tracy (0.4.3) {84341 KB} -> 044/asset-new-id.tracy (0.4.4) {81771 KB} 96.95% size change 043/asset-old.tracy (0.4.3) {80688 KB} -> 044/asset-old.tracy (0.4.4) {78410 KB} 97.18% size change 043/big.tracy (0.4.3) {939577 KB} -> 044/big.tracy (0.4.4) {938427 KB} 99.88% size change 043/callstack.tracy (0.4.3) {14557 KB} -> 044/callstack.tracy (0.4.4) {14465 KB} 99.37% size change 043/callstack-linux.tracy (0.4.3) {6949 KB} -> 044/callstack-linux.tracy (0.4.4) {6942 KB} 99.90% size change 043/crash.tracy (0.4.3) {131 KB} -> 044/crash.tracy (0.4.4) {127 KB} 97.10% size change 043/crash2.tracy (0.4.3) {1422 KB} -> 044/crash2.tracy (0.4.4) {1412 KB} 99.29% size change 043/darkrl.tracy (0.4.3) {15767 KB} -> 044/darkrl.tracy (0.4.4) {15663 KB} 99.34% size change 043/darkrl2.tracy (0.4.3) {7947 KB} -> 044/darkrl2.tracy (0.4.4) {7886 KB} 99.23% size change 043/darkrl-old.tracy (0.4.3) {67448 KB} -> 044/darkrl-old.tracy (0.4.4) {67004 KB} 99.34% size change 043/deadlock.tracy (0.4.3) {5984 KB} -> 044/deadlock.tracy (0.4.4) {5986 KB} 100.03% size change 043/gn-opengl.tracy (0.4.3) {29005 KB} -> 044/gn-opengl.tracy (0.4.4) {28885 KB} 99.59% size change 043/gn-vulkan.tracy (0.4.3) {29352 KB} -> 044/gn-vulkan.tracy (0.4.4) {29257 KB} 99.68% size change 043/long.tracy (0.4.3) {1182800 KB} -> 044/long.tracy (0.4.4) {1176584 KB} 99.47% size change 043/mem.tracy (0.4.3) {1369067 KB} -> 044/mem.tracy (0.4.4) {1262406 KB} 92.21% size change 043/multi.tracy (0.4.3) {8004 KB} -> 044/multi.tracy (0.4.4) {7944 KB} 99.24% size change 043/new.tracy (0.4.3) {1108 KB} -> 044/new.tracy (0.4.4) {1099 KB} 99.18% size change 043/q3bsp-mt.tracy (0.4.3) {949855 KB} -> 044/q3bsp-mt.tracy (0.4.4) {937574 KB} 98.71% size change 043/q3bsp-st.tracy (0.4.3) {240347 KB} -> 044/q3bsp-st.tracy (0.4.4) {230092 KB} 95.73% size change 043/selfprofile.tracy (0.4.3) {197708 KB} -> 044/selfprofile.tracy (0.4.4) {197659 KB} 99.98% size change 043/tbrowser.tracy (0.4.3) {9503 KB} -> 044/tbrowser.tracy (0.4.4) {9503 KB} 100.00% size change 043/test.tracy (0.4.3) {40700 KB} -> 044/test.tracy (0.4.4) {40699 KB} 100.00% size change 043/virtualfile_hc.tracy (0.4.3) {72424 KB} -> 044/virtualfile_hc.tracy (0.4.4) {72304 KB} 99.83% size change 043/zfile_hc.tracy (0.4.3) {39419 KB} -> 044/zfile_hc.tracy (0.4.4) {39328 KB} 99.77% size change
2019-02-16 22:04:34 +00:00
tracy_force_inline void ReadTimelinePre044( FileRead& f, GpuEvent* zone, int64_t& refTime, int64_t& refGpuTime, int fileVer );
Don't call ReadTimeline() when there's nothing to read.
2018-03-15 21:54:10 +00:00
Take callstack field in ZoneEvent into account in save/load.
2018-06-21 23:30:08 +00:00
tracy_force_inline void ReadTimelineUpdateStatistics( ZoneEvent* zone, uint16_t thread );
Store time deltas, instead of absolute time in trace dumps. This change greatly reduces the size of saved dumps, but increase the cost of processing during loading. One notable outlier in the dataset below is mem.tracy, which increased in size, even if changes in the memory dump saving scheme decrease size of the other traces. 041/aa.tracy (0.4.1) {18987 KB} -> 042/aa.tracy (0.4.2) {10140 KB} 53.40% size change 041/android.tracy (0.4.1) {696753 KB} -> 042/android.tracy (0.4.2) {542738 KB} 77.90% size change 041/asset-new.tracy (0.4.1) {97163 KB} -> 042/asset-new.tracy (0.4.2) {78402 KB} 80.69% size change 041/asset-new-id.tracy (0.4.1) {105683 KB} -> 042/asset-new-id.tracy (0.4.2) {84341 KB} 79.81% size change 041/asset-old.tracy (0.4.1) {100205 KB} -> 042/asset-old.tracy (0.4.2) {80688 KB} 80.52% size change 041/big.tracy (0.4.1) {2246014 KB} -> 042/big.tracy (0.4.2) {943083 KB} 41.99% size change 041/crash.tracy (0.4.1) {143 KB} -> 042/crash.tracy (0.4.2) {131 KB} 91.39% size change 041/crash2.tracy (0.4.1) {3411 KB} -> 042/crash2.tracy (0.4.2) {1425 KB} 41.80% size change 041/darkrl.tracy (0.4.1) {31818 KB} -> 042/darkrl.tracy (0.4.2) {15897 KB} 49.96% size change 041/darkrl2.tracy (0.4.1) {18778 KB} -> 042/darkrl2.tracy (0.4.2) {8002 KB} 42.62% size change 041/darkrl-old.tracy (0.4.1) {151346 KB} -> 042/darkrl-old.tracy (0.4.2) {67945 KB} 44.89% size change 041/deadlock.tracy (0.4.1) {53 KB} -> 042/deadlock.tracy (0.4.2) {52 KB} 98.55% size change 041/gn-opengl.tracy (0.4.1) {45860 KB} -> 042/gn-opengl.tracy (0.4.2) {30983 KB} 67.56% size change 041/gn-vulkan.tracy (0.4.1) {45618 KB} -> 042/gn-vulkan.tracy (0.4.2) {31349 KB} 68.72% size change 041/long.tracy (0.4.1) {1583550 KB} -> 042/long.tracy (0.4.2) {1225316 KB} 77.38% size change 041/mem.tracy (0.4.1) {1243058 KB} -> 042/mem.tracy (0.4.2) {1369291 KB} 110.15% size change 041/multi.tracy (0.4.1) {14519 KB} -> 042/multi.tracy (0.4.2) {8110 KB} 55.86% size change 041/new.tracy (0.4.1) {1439 KB} -> 042/new.tracy (0.4.2) {1108 KB} 77.01% size change 041/q3bsp-mt.tracy (0.4.1) {1414323 KB} -> 042/q3bsp-mt.tracy (0.4.2) {949855 KB} 67.16% size change 041/q3bsp-st.tracy (0.4.1) {301334 KB} -> 042/q3bsp-st.tracy (0.4.2) {240347 KB} 79.76% size change 041/selfprofile.tracy (0.4.1) {399648 KB} -> 042/selfprofile.tracy (0.4.2) {197713 KB} 49.47% size change 041/tbrowser.tracy (0.4.1) {13052 KB} -> 042/tbrowser.tracy (0.4.2) {9503 KB} 72.81% size change 041/test.tracy (0.4.1) {60309 KB} -> 042/test.tracy (0.4.2) {40700 KB} 67.49% size change 041/virtualfile_hc.tracy (0.4.1) {108967 KB} -> 042/virtualfile_hc.tracy (0.4.2) {72839 KB} 66.85% size change 041/zfile_hc.tracy (0.4.1) {58814 KB} -> 042/zfile_hc.tracy (0.4.2) {39608 KB} 67.35% size change
2018-12-30 22:06:03 +00:00
void ReadTimeline( FileRead& f, Vector<ZoneEvent*>& vec, uint16_t thread, uint64_t size, int64_t& refTime );
void ReadTimelinePre042( FileRead& f, Vector<ZoneEvent*>& vec, uint16_t thread, uint64_t size, int fileVer );
Track separate time offset for GPU times. This is second version of 0.4.2 dump file format. Previous 0.4.2 format cannot be read anymore. 041/aa.tracy (0.4.1) {18987 KB} -> 042/aa.tracy (0.4.2) {10051 KB} 52.94% size change 041/android.tracy (0.4.1) {696753 KB} -> 042/android.tracy (0.4.2) {542738 KB} 77.90% size change 041/asset-new.tracy (0.4.1) {97163 KB} -> 042/asset-new.tracy (0.4.2) {78402 KB} 80.69% size change 041/asset-new-id.tracy (0.4.1) {105683 KB} -> 042/asset-new-id.tracy (0.4.2) {84341 KB} 79.81% size change 041/asset-old.tracy (0.4.1) {100205 KB} -> 042/asset-old.tracy (0.4.2) {80688 KB} 80.52% size change 041/big.tracy (0.4.1) {2246014 KB} -> 042/big.tracy (0.4.2) {939578 KB} 41.83% size change 041/crash.tracy (0.4.1) {143 KB} -> 042/crash.tracy (0.4.2) {131 KB} 91.37% size change 041/crash2.tracy (0.4.1) {3411 KB} -> 042/crash2.tracy (0.4.2) {1420 KB} 41.63% size change 041/darkrl.tracy (0.4.1) {31818 KB} -> 042/darkrl.tracy (0.4.2) {15762 KB} 49.54% size change 041/darkrl2.tracy (0.4.1) {18778 KB} -> 042/darkrl2.tracy (0.4.2) {7945 KB} 42.31% size change 041/darkrl-old.tracy (0.4.1) {151346 KB} -> 042/darkrl-old.tracy (0.4.2) {67449 KB} 44.57% size change 041/deadlock.tracy (0.4.1) {53 KB} -> 042/deadlock.tracy (0.4.2) {52 KB} 98.55% size change 041/gn-opengl.tracy (0.4.1) {45860 KB} -> 042/gn-opengl.tracy (0.4.2) {29005 KB} 63.25% size change 041/gn-vulkan.tracy (0.4.1) {45618 KB} -> 042/gn-vulkan.tracy (0.4.2) {29352 KB} 64.34% size change 041/long.tracy (0.4.1) {1583550 KB} -> 042/long.tracy (0.4.2) {1182800 KB} 74.69% size change 041/mem.tracy (0.4.1) {1243058 KB} -> 042/mem.tracy (0.4.2) {1369067 KB} 110.14% size change 041/multi.tracy (0.4.1) {14519 KB} -> 042/multi.tracy (0.4.2) {8000 KB} 55.10% size change 041/new.tracy (0.4.1) {1439 KB} -> 042/new.tracy (0.4.2) {1105 KB} 76.75% size change 041/q3bsp-mt.tracy (0.4.1) {1414323 KB} -> 042/q3bsp-mt.tracy (0.4.2) {949855 KB} 67.16% size change 041/q3bsp-st.tracy (0.4.1) {301334 KB} -> 042/q3bsp-st.tracy (0.4.2) {240347 KB} 79.76% size change 041/selfprofile.tracy (0.4.1) {399648 KB} -> 042/selfprofile.tracy (0.4.2) {197704 KB} 49.47% size change 041/tbrowser.tracy (0.4.1) {13052 KB} -> 042/tbrowser.tracy (0.4.2) {9503 KB} 72.81% size change 041/test.tracy (0.4.1) {60309 KB} -> 042/test.tracy (0.4.2) {40700 KB} 67.49% size change 041/virtualfile_hc.tracy (0.4.1) {108967 KB} -> 042/virtualfile_hc.tracy (0.4.2) {72424 KB} 66.46% size change 041/zfile_hc.tracy (0.4.1) {58814 KB} -> 042/zfile_hc.tracy (0.4.2) {39418 KB} 67.02% size change
2019-01-03 19:09:45 +00:00
void ReadTimeline( FileRead& f, Vector<GpuEvent*>& vec, uint64_t size, int64_t& refTime, int64_t& refGpuTime );
Don't decompress GpuZone threads while saving trace. Saving the threads compressed in GPU zones and memory event has the following result on trace dump sizes: 043/aa.tracy (0.4.3) {10055 KB} -> 044/aa.tracy (0.4.4) {9975 KB} 99.20% size change 043/android.tracy (0.4.3) {542739 KB} -> 044/android.tracy (0.4.4) {519248 KB} 95.67% size change 043/asset-new.tracy (0.4.3) {78403 KB} -> 044/asset-new.tracy (0.4.4) {75899 KB} 96.81% size change 043/asset-new-id.tracy (0.4.3) {84341 KB} -> 044/asset-new-id.tracy (0.4.4) {81771 KB} 96.95% size change 043/asset-old.tracy (0.4.3) {80688 KB} -> 044/asset-old.tracy (0.4.4) {78410 KB} 97.18% size change 043/big.tracy (0.4.3) {939577 KB} -> 044/big.tracy (0.4.4) {938427 KB} 99.88% size change 043/callstack.tracy (0.4.3) {14557 KB} -> 044/callstack.tracy (0.4.4) {14465 KB} 99.37% size change 043/callstack-linux.tracy (0.4.3) {6949 KB} -> 044/callstack-linux.tracy (0.4.4) {6942 KB} 99.90% size change 043/crash.tracy (0.4.3) {131 KB} -> 044/crash.tracy (0.4.4) {127 KB} 97.10% size change 043/crash2.tracy (0.4.3) {1422 KB} -> 044/crash2.tracy (0.4.4) {1412 KB} 99.29% size change 043/darkrl.tracy (0.4.3) {15767 KB} -> 044/darkrl.tracy (0.4.4) {15663 KB} 99.34% size change 043/darkrl2.tracy (0.4.3) {7947 KB} -> 044/darkrl2.tracy (0.4.4) {7886 KB} 99.23% size change 043/darkrl-old.tracy (0.4.3) {67448 KB} -> 044/darkrl-old.tracy (0.4.4) {67004 KB} 99.34% size change 043/deadlock.tracy (0.4.3) {5984 KB} -> 044/deadlock.tracy (0.4.4) {5986 KB} 100.03% size change 043/gn-opengl.tracy (0.4.3) {29005 KB} -> 044/gn-opengl.tracy (0.4.4) {28885 KB} 99.59% size change 043/gn-vulkan.tracy (0.4.3) {29352 KB} -> 044/gn-vulkan.tracy (0.4.4) {29257 KB} 99.68% size change 043/long.tracy (0.4.3) {1182800 KB} -> 044/long.tracy (0.4.4) {1176584 KB} 99.47% size change 043/mem.tracy (0.4.3) {1369067 KB} -> 044/mem.tracy (0.4.4) {1262406 KB} 92.21% size change 043/multi.tracy (0.4.3) {8004 KB} -> 044/multi.tracy (0.4.4) {7944 KB} 99.24% size change 043/new.tracy (0.4.3) {1108 KB} -> 044/new.tracy (0.4.4) {1099 KB} 99.18% size change 043/q3bsp-mt.tracy (0.4.3) {949855 KB} -> 044/q3bsp-mt.tracy (0.4.4) {937574 KB} 98.71% size change 043/q3bsp-st.tracy (0.4.3) {240347 KB} -> 044/q3bsp-st.tracy (0.4.4) {230092 KB} 95.73% size change 043/selfprofile.tracy (0.4.3) {197708 KB} -> 044/selfprofile.tracy (0.4.4) {197659 KB} 99.98% size change 043/tbrowser.tracy (0.4.3) {9503 KB} -> 044/tbrowser.tracy (0.4.4) {9503 KB} 100.00% size change 043/test.tracy (0.4.3) {40700 KB} -> 044/test.tracy (0.4.4) {40699 KB} 100.00% size change 043/virtualfile_hc.tracy (0.4.3) {72424 KB} -> 044/virtualfile_hc.tracy (0.4.4) {72304 KB} 99.83% size change 043/zfile_hc.tracy (0.4.3) {39419 KB} -> 044/zfile_hc.tracy (0.4.4) {39328 KB} 99.77% size change
2019-02-16 22:04:34 +00:00
void ReadTimelinePre044( FileRead& f, Vector<GpuEvent*>& vec, uint64_t size, int64_t& refTime, int64_t& refGpuTime, int fileVer );
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
Store time deltas, instead of absolute time in trace dumps. This change greatly reduces the size of saved dumps, but increase the cost of processing during loading. One notable outlier in the dataset below is mem.tracy, which increased in size, even if changes in the memory dump saving scheme decrease size of the other traces. 041/aa.tracy (0.4.1) {18987 KB} -> 042/aa.tracy (0.4.2) {10140 KB} 53.40% size change 041/android.tracy (0.4.1) {696753 KB} -> 042/android.tracy (0.4.2) {542738 KB} 77.90% size change 041/asset-new.tracy (0.4.1) {97163 KB} -> 042/asset-new.tracy (0.4.2) {78402 KB} 80.69% size change 041/asset-new-id.tracy (0.4.1) {105683 KB} -> 042/asset-new-id.tracy (0.4.2) {84341 KB} 79.81% size change 041/asset-old.tracy (0.4.1) {100205 KB} -> 042/asset-old.tracy (0.4.2) {80688 KB} 80.52% size change 041/big.tracy (0.4.1) {2246014 KB} -> 042/big.tracy (0.4.2) {943083 KB} 41.99% size change 041/crash.tracy (0.4.1) {143 KB} -> 042/crash.tracy (0.4.2) {131 KB} 91.39% size change 041/crash2.tracy (0.4.1) {3411 KB} -> 042/crash2.tracy (0.4.2) {1425 KB} 41.80% size change 041/darkrl.tracy (0.4.1) {31818 KB} -> 042/darkrl.tracy (0.4.2) {15897 KB} 49.96% size change 041/darkrl2.tracy (0.4.1) {18778 KB} -> 042/darkrl2.tracy (0.4.2) {8002 KB} 42.62% size change 041/darkrl-old.tracy (0.4.1) {151346 KB} -> 042/darkrl-old.tracy (0.4.2) {67945 KB} 44.89% size change 041/deadlock.tracy (0.4.1) {53 KB} -> 042/deadlock.tracy (0.4.2) {52 KB} 98.55% size change 041/gn-opengl.tracy (0.4.1) {45860 KB} -> 042/gn-opengl.tracy (0.4.2) {30983 KB} 67.56% size change 041/gn-vulkan.tracy (0.4.1) {45618 KB} -> 042/gn-vulkan.tracy (0.4.2) {31349 KB} 68.72% size change 041/long.tracy (0.4.1) {1583550 KB} -> 042/long.tracy (0.4.2) {1225316 KB} 77.38% size change 041/mem.tracy (0.4.1) {1243058 KB} -> 042/mem.tracy (0.4.2) {1369291 KB} 110.15% size change 041/multi.tracy (0.4.1) {14519 KB} -> 042/multi.tracy (0.4.2) {8110 KB} 55.86% size change 041/new.tracy (0.4.1) {1439 KB} -> 042/new.tracy (0.4.2) {1108 KB} 77.01% size change 041/q3bsp-mt.tracy (0.4.1) {1414323 KB} -> 042/q3bsp-mt.tracy (0.4.2) {949855 KB} 67.16% size change 041/q3bsp-st.tracy (0.4.1) {301334 KB} -> 042/q3bsp-st.tracy (0.4.2) {240347 KB} 79.76% size change 041/selfprofile.tracy (0.4.1) {399648 KB} -> 042/selfprofile.tracy (0.4.2) {197713 KB} 49.47% size change 041/tbrowser.tracy (0.4.1) {13052 KB} -> 042/tbrowser.tracy (0.4.2) {9503 KB} 72.81% size change 041/test.tracy (0.4.1) {60309 KB} -> 042/test.tracy (0.4.2) {40700 KB} 67.49% size change 041/virtualfile_hc.tracy (0.4.1) {108967 KB} -> 042/virtualfile_hc.tracy (0.4.2) {72839 KB} 66.85% size change 041/zfile_hc.tracy (0.4.1) {58814 KB} -> 042/zfile_hc.tracy (0.4.2) {39608 KB} 67.35% size change
2018-12-30 22:06:03 +00:00
void WriteTimeline( FileWrite& f, const Vector<ZoneEvent*>& vec, int64_t& refTime );
Track separate time offset for GPU times. This is second version of 0.4.2 dump file format. Previous 0.4.2 format cannot be read anymore. 041/aa.tracy (0.4.1) {18987 KB} -> 042/aa.tracy (0.4.2) {10051 KB} 52.94% size change 041/android.tracy (0.4.1) {696753 KB} -> 042/android.tracy (0.4.2) {542738 KB} 77.90% size change 041/asset-new.tracy (0.4.1) {97163 KB} -> 042/asset-new.tracy (0.4.2) {78402 KB} 80.69% size change 041/asset-new-id.tracy (0.4.1) {105683 KB} -> 042/asset-new-id.tracy (0.4.2) {84341 KB} 79.81% size change 041/asset-old.tracy (0.4.1) {100205 KB} -> 042/asset-old.tracy (0.4.2) {80688 KB} 80.52% size change 041/big.tracy (0.4.1) {2246014 KB} -> 042/big.tracy (0.4.2) {939578 KB} 41.83% size change 041/crash.tracy (0.4.1) {143 KB} -> 042/crash.tracy (0.4.2) {131 KB} 91.37% size change 041/crash2.tracy (0.4.1) {3411 KB} -> 042/crash2.tracy (0.4.2) {1420 KB} 41.63% size change 041/darkrl.tracy (0.4.1) {31818 KB} -> 042/darkrl.tracy (0.4.2) {15762 KB} 49.54% size change 041/darkrl2.tracy (0.4.1) {18778 KB} -> 042/darkrl2.tracy (0.4.2) {7945 KB} 42.31% size change 041/darkrl-old.tracy (0.4.1) {151346 KB} -> 042/darkrl-old.tracy (0.4.2) {67449 KB} 44.57% size change 041/deadlock.tracy (0.4.1) {53 KB} -> 042/deadlock.tracy (0.4.2) {52 KB} 98.55% size change 041/gn-opengl.tracy (0.4.1) {45860 KB} -> 042/gn-opengl.tracy (0.4.2) {29005 KB} 63.25% size change 041/gn-vulkan.tracy (0.4.1) {45618 KB} -> 042/gn-vulkan.tracy (0.4.2) {29352 KB} 64.34% size change 041/long.tracy (0.4.1) {1583550 KB} -> 042/long.tracy (0.4.2) {1182800 KB} 74.69% size change 041/mem.tracy (0.4.1) {1243058 KB} -> 042/mem.tracy (0.4.2) {1369067 KB} 110.14% size change 041/multi.tracy (0.4.1) {14519 KB} -> 042/multi.tracy (0.4.2) {8000 KB} 55.10% size change 041/new.tracy (0.4.1) {1439 KB} -> 042/new.tracy (0.4.2) {1105 KB} 76.75% size change 041/q3bsp-mt.tracy (0.4.1) {1414323 KB} -> 042/q3bsp-mt.tracy (0.4.2) {949855 KB} 67.16% size change 041/q3bsp-st.tracy (0.4.1) {301334 KB} -> 042/q3bsp-st.tracy (0.4.2) {240347 KB} 79.76% size change 041/selfprofile.tracy (0.4.1) {399648 KB} -> 042/selfprofile.tracy (0.4.2) {197704 KB} 49.47% size change 041/tbrowser.tracy (0.4.1) {13052 KB} -> 042/tbrowser.tracy (0.4.2) {9503 KB} 72.81% size change 041/test.tracy (0.4.1) {60309 KB} -> 042/test.tracy (0.4.2) {40700 KB} 67.49% size change 041/virtualfile_hc.tracy (0.4.1) {108967 KB} -> 042/virtualfile_hc.tracy (0.4.2) {72424 KB} 66.46% size change 041/zfile_hc.tracy (0.4.1) {58814 KB} -> 042/zfile_hc.tracy (0.4.2) {39418 KB} 67.02% size change
2019-01-03 19:09:45 +00:00
void WriteTimeline( FileWrite& f, const Vector<GpuEvent*>& vec, int64_t& refTime, int64_t& refGpuTime );
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
int64_t TscTime( int64_t tsc ) { return int64_t( tsc * m_timerMul ); }
int64_t TscTime( uint64_t tsc ) { return int64_t( tsc * m_timerMul ); }
Socket m_sock;
std::string m_addr;
Sort source location zones on a separate thread.
2018-04-30 01:54:09 +00:00
std::thread m_thread;
Fix initialization of atomics.
2019-01-06 20:09:56 +00:00
std::atomic<bool> m_connected = { false };
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
std::atomic<bool> m_hasData;
Fix initialization of atomics.
2019-01-06 20:09:56 +00:00
std::atomic<bool> m_shutdown = { false };
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
Improve handling of post-load background jobs. Background tasks (source location zones sorting, reconstruction of memory plot) are now started only after trace loading is finished. Multithreaded sorting was previously impacting trace load times. Only one thread is used to perform both jobs, one after another.
2019-02-14 00:17:37 +00:00
std::thread m_threadBackground;
Sort source location zones on a separate thread.
2018-04-30 01:54:09 +00:00
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
int64_t m_delay;
int64_t m_resolution;
double m_timerMul;
std::string m_captureName;
Store explicit program name and capture time.
2018-08-28 22:57:11 +00:00
std::string m_captureProgram;
uint64_t m_captureTime;
Store host info.
2018-08-19 16:21:56 +00:00
std::string m_hostInfo;
Init common Worker variables in header.
2019-01-06 18:04:50 +00:00
bool m_terminate = false;
bool m_crashed = false;
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
LZ4_streamDecode_t* m_stream;
char* m_buffer;
int m_bufferOffset;
Store on-demand connection state.
2018-07-11 23:21:04 +00:00
bool m_onDemand;
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
Don't use hash table to store 256 pointers.
2018-06-22 13:14:44 +00:00
GpuCtxData* m_gpuCtxMap[256];
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
flat_hash_map<uint64_t, StringLocation, nohash<uint64_t>> m_pendingCustomStrings;
Only one callstack may be in-flight at any time. Save for the allocated callstack, but this will be solved in another way. There's no need to keep pending callstacks in a map.
2019-03-05 18:30:17 +00:00
uint64_t m_pendingCallstackPtr = 0;
uint32_t m_pendingCallstackId;
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
flat_hash_map<uint64_t, int32_t, nohash<uint64_t>> m_pendingSourceLocationPayload;
Vector<uint64_t> m_sourceLocationQueue;
flat_hash_map<uint64_t, uint32_t, nohash<uint64_t>> m_sourceLocationShrink;
flat_hash_map<uint64_t, ThreadData*, nohash<uint64_t>> m_threadMap;
Handle ZoneBeginCallstack queue event. This is identical to ZoneBegin handling, but requires some additional bookkeeping to account for the incoming callstack information.
2018-06-21 23:07:25 +00:00
flat_hash_map<uint64_t, NextCallstack, nohash<uint64_t>> m_nextCallstack;
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
uint32_t m_pendingStrings;
uint32_t m_pendingThreads;
uint32_t m_pendingSourceLocation;
Fix typo.
2018-06-20 21:42:00 +00:00
uint32_t m_pendingCallstackFrames;
Make it possible to store multiple frames at single frame address.
2019-01-20 18:11:48 +00:00
uint8_t m_pendingCallstackSubframes;
CallstackFrameData* m_callstackFrameStaging;
uint64_t m_callstackFrameStagingPtr;
Store callstack alloc payloads.
2019-03-03 16:34:56 +00:00
uint64_t m_callstackAllocNextIdx = 0;
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
Process memory events with callstack.
2018-06-19 16:52:45 +00:00
uint64_t m_lastMemActionCallstack;
Have to track callstacks of both alloc and free.
2018-06-19 20:08:47 +00:00
bool m_lastMemActionWasAlloc;
Process memory events with callstack.
2018-06-19 16:52:45 +00:00
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
Slab<64*1024*1024> m_slab;
DataBlock m_data;
MbpsBlock m_mbpsData;
Track trace loading progress.
2018-07-28 15:59:17 +00:00
Track trace version in worker.
2018-07-29 13:33:48 +00:00
int m_traceVersion;
Fix initialization of atomics.
2019-01-06 20:09:56 +00:00
std::atomic<uint8_t> m_handshake = { 0 };
Track trace version in worker.
2018-07-29 13:33:48 +00:00
Track trace loading progress.
2018-07-28 15:59:17 +00:00
static LoadProgress s_loadProgress;
Track trace loading time.
2019-01-06 18:09:50 +00:00
int64_t m_loadTime;
Gracefully store failure states.
2019-01-14 22:22:31 +00:00
Failure m_failure = Failure::None;
FailureData m_failureData;
Process CPU usage reports.
2019-02-21 21:45:39 +00:00
PlotData* m_sysTimePlot = nullptr;
Split data handling code from the view.
2018-02-13 13:57:47 +00:00
};
}
#endif
Reference in New Issue Copy Permalink