Age | Commit message (Collapse) | Author |
|
made getPrimaryAccumulator return a reference since it was an
always non-null pointer
changed unit conversion to perform lazy division in order to avoid truncation
of timer values
|
|
|
|
added convenience types for units F32Seconds, etc.
|
|
removed LLThreadLocalSingleton
collapsed all thread recorder classes to single type, LLTrace::ThreadRecorder
moved fasttimer stack head to llthreadlocalsingletonpointer via ThreadRecorder
|
|
console
fixed some lltrace logic errors
more consistent syncing of timestamps of sample values in recording stack
selection of primary buffers was completely incorrect
assignment of recordings got wrong play state due to implicit
operator = defined in base class
fixed asset stats only working up to the first send
|
|
moved collapsed flag to fast timer tree node
|
|
removed implicit flushes on reads from recorders for better performance
made sure stack timers were updated on recorder deactivate
faster rendering and better ui for fast timer view
|
|
changed Units macros and argument order to make it more clear
optimized units for integer types
fixed merging of periodicrecordings...should eliminate duplicate entries in sceneloadmonitor history
|
|
|
|
fixed mem stat tracking...now properly tracks memory footprint with floating point
precision
cleaned up macros for unit declaration
renamed units to SI standard for 1024 multiples (kibibytes, etc)
fixed units output for scene monitor dump
|
|
|
|
|
|
fix for inaccurate optimization of full block time calculations
|
|
streamlined fast timer delta tracking
|
|
fixed anamolous LLFastTimer timings
|
|
|
|
added debug output to BlockTimer
|
|
fix for timings for recursive fast timers not being correct
|
|
A couple of merge issues that caused the resulting code to not build.
|
|
fixed crash on startup
|
|
fixed some uninitialized variables
root timer accumulator was being initialized to NULL
|
|
improved performance of fast timer stat gathering
|
|
|
|
made fast timer stack thread local
added LLThreadLocalSingleton
made LLThreadLocalPointer obey pointer rules for const
added LLThreadLocalSingletonPointer for fast thread local pointers
|
|
improvements on lifetime of lltrace core data structures
tweaks to thread local pointer handling so that static constructors/destructors
can safely call functions that use lltrace
|
|
removed some potential data races
got memory stats recording in trace system
|
|
added ability to query self time of block timers indepedently
|
|
converted fast timer view over to new lltrace mechanisms
|
|
moved runtime timer tree topology information to separate array instead of recording stack
|
|
hunting down bad values and crashes
|
|
eliminated min and max macros from windows.h
got rest of viewer to compile against llfasttimer changes
|
|
cleaning up build
moved most includes of windows.h to llwin32headers.h to disable min/max macros, etc
streamlined Time class and consolidated functionality in BlockTimer class
llfasttimer is no longer included via llstring.h, so had to add it manually in several places
|
|
moving fast timers into lltrace namespace and accumulation system
|
|
removed remnants of LLFastTimer::FrameState
|
|
simplified llfasttimer code down to 2 classes
llunit unit conversion now done in floating point or 64 bit integer precision,
depending on source type
|
|
build fix
|
|
first pass at LLTrace framework
|
|
|
|
cleaner implementation of llfasttimers...don't bother to share similarly named timers
just create multiple timers with same name...doesn't break anything
|
|
allow duplicate named fast timers again, refactored timer code
|
|
removed unnecessary cache miss from fast timers
renamed llfasttimer_class back to llfasttimer
|
|
|
|
|
|
|
|
|
|
|
|
|
|
syscall was REALLY chewing CPU time. Sigh. I didn't realize how incredibly often this gets called. So, back to the assembly.
But be more careful with CPU clock count on linux, so the fasttimer values are much more accurate than they were the last time we were with RDTSC, in absolute terms - back in the right order of magnitude anyway.
Also change many instances of Mhz to MHz.
Also some minor comment fixes.
|
|
more reliable fix based on feedback from Richard. dicked with the Darwin results too since those seemed wrong based on the same feedback (also covered in test plan).
|
|
but the compiler didn't start caring until I made these proper member functions.
fixed.
|