Optimizing Multithreaded Simulation Regions — Handmade Hero — Episode Guide

0:35A few words on the new format

1:41Recap and set the stage for the day

2:43Run the game, consult the profiler and assess our current situation

9:32handmade_entity.cpp: Stop profiling the entity system, run the game and consult the event count

11:48handmade_world.cpp: Stop profiling GetWorldChunkInternal(), run the game and consult the profiler

13:22handmade_world_mode.cpp: Make PlayWorld() set the WorldChunkDimInMeters larger, run the game and consider drawing the bounds of those chunks

16:46handmade_world.cpp: Introduce GetWorldChunkBounds()

21:25handmade_world_mode.cpp: Make UpdateAndRenderWorld() call GetWorldChunkBounds() and draw them

23:22Run the game and view that debug visualisation for the chunk

26:52handmade_world_mode.cpp: Make PlayWorld() set the chunk size roughly equal to the room size, run the game and view the visualisation

29:31handmade_sim_region.cpp: Change BeginWorldChange() to only simulate one room's worth

31:20Run the game and note how our simulation region erroneously splits entities

32:29handmade_world_mode.cpp: Make UpdateAndRenderWorld() operate on the eight rooms above, below and around the player

36:33Run the game to take a look at how that's working

38:46handmade_platform.h: Add BeginTicketMutex() to the profiler, run the game to consult the profiler and consider how to improve the multithreading

42:11handmade_sim_region.cpp: Enable BeginWorldChange() to do a bunch of work without needing to be in a mutex

46:17handmade_world.cpp: Enable AddToFreeList() to insert the whole list in one go

48:05Run the game and consult the profiler

50:27handmade_world.cpp: Make RemoveWorldChunk() and AddToFreeList() themselves begin and end the mutex

52:54handmade_world.cpp: Consider what UseChunkSpace() does, make it begin and end a mutex, run the game and consult the profiler

56:22handmade_sim_region.cpp: Consider how to improve the ticket-taking in BeginWorldChange()

59:38Run the game and consult the profiler until we hit a deadlock and investigate why

1:02:43handmade_world.cpp: Make the correct version of UseChunkSpace() begin and end a mutex, run the game and consult the profiler again

1:04:40handmade_world.cpp: Add RemoveWorldChunk() and AddToFreeList() to the profiler, run the game and consult it

1:06:00handmade_sim_region.cpp: Add more timing blocks and functions to the profiler, running the game and consulting that profiler

1:10:04handmade_sim_region.cpp: Establish that the PushStruct() call in BeginWorldChange() is taking half our time

1:10:37Note that the timer totally failed

1:11:09handmade_memory.h: Investigate why that PushStruct() is so expensive

1:13:11handmade_sim_region.cpp: Add a timing block around that PushStruct() call in BeginWorldChange(), run the game and inspect the profiler

1:14:11Consider why we are clearing this buffer

1:16:20Break

1:21:35Return and wonder if the chat figured out what's wrong with the code

1:23:30handmade_sim_region.cpp: Make BeginWorldChange() pass NoClear to that PushStruct() call and run the game to consult the profiler

1:25:11handmade_sim_region.cpp: Time the ZeroStruct() call, run the game to consult the profiler and consider how to optimise this routine

1:28:27handmade_debug_ui.cpp: Enable DrawTopClocksList() to print out the cycles per invocation

1:29:58handmade_math.h: Introduce a 64-bit version of SafeRatio0()

1:30:24Run the game and consult the profiler

1:31:12Break into the ZeroStruct() call in BeginWorldChange() and determine that the clear is costing 7 cycles per byte

1:33:55handmade_sim_region.cpp: Verify that the clear's cycle-time is reasonable

1:37:21Consider ways to minimise the memory footprint

1:40:55handmade_sim_region.h: Add EntityHashOccupancy and BrainHashOccupancy to the sim_region struct in order to enable doing the clear with a bitfield

1:44:03handmade_sim_region.cpp: Introduce MarkOccupied(), MarkBit() and IsEmpty()

1:49:03Run the game, hit the assert in AddEntityToHash(), and fix that assertion to handle the new scheme

1:51:12handmade_sim_region.cpp: Fix the sense of IsEmpty(), run the game and crash in ExecuteBrain()

1:52:02Build and run in -Od and investigate what's happening

1:54:18handmade_sim_region.cpp: Add an assert in AddEntityToHash() to validate the hash, run the game and do not hit that assertion

1:55:07Step through AddEntityToHash() and investigate what's happening

1:56:30handmade_sim_region.cpp: Walk through GetOrAddBrain()

1:57:38handmade_sim_region.cpp: Make GetOrAddBrain() set the Hash->ID, run the game and crash in FreeFrame()

1:58:31handmade_sim_region.h: Remove the ID from the entity_hash and brain_hash and instead use their Ptr to identify them

1:59:36Run the game and note that "our brains are still not working properly"^α

2:03:55handmade_sim_region.cpp: Add assertions to GetOrAddBrain() in order to verify ID resolution

2:06:51handmade_sim_region.cpp: Make BeginWorldChange() and GetOrAddBrain() clear the Brains, run the game and note that we're still getting weird behaviour

2:08:06handmade_memory.h: Enable overflow checking in PushSize_(), run the game and...

2:08:41"So we don't seem to be having any overflow problems at the moment... *stream crashes*"^β

2:09:43Wait for Windows to decide to suspend the process

2:13:23Run the game normally having removed that overflow checking, and watch the memory counter

2:14:30handmade_sim_region.cpp: Make BeginWorldChange() clear the SimRegion, run the game and watch the profiler

2:15:40handmade_sim_region.cpp: Temporarily make BeginWorldChange() clear the EntityHash and BrainHash, run the game and watch the profiler

2:17:39handmade_sim_region.cpp: Make GetHashFromID() clear upon encountering an empty Entry for now

2:18:42Run the game successfully and consult the profiler

2:19:52Q&A

🗩

2:19:52Q&A

🗩

2:19:52Q&A

🗩

2:20:17AsafGartner The stream is down

🗪

2:20:17AsafGartner The stream is down

🗪

2:20:17AsafGartner The stream is down

🗪

2:22:51AsafGartner Did you fix the bug?

🗪

2:22:51AsafGartner Did you fix the bug?

🗪

2:22:51AsafGartner Did you fix the bug?

🗪

2:23:45AsafGartner There's a TODO in ZeroSize. Not sure if you noticed it

🗪

2:23:45AsafGartner There's a TODO in ZeroSize. Not sure if you noticed it

🗪

2:23:45AsafGartner There's a TODO in ZeroSize. Not sure if you noticed it

🗪

2:24:07handmade_memory.h: Remove the TODO from ZeroSize()

2:24:23uplinkcoder What was the bug?

🗪

2:24:23uplinkcoder What was the bug?

🗪

2:24:23uplinkcoder What was the bug?

🗪

2:25:26longboolean We should have something built into the build script that won't let you compile if the stream has gone down. Is this possible to do in a batch script?

🗪

2:25:26longboolean We should have something built into the build script that won't let you compile if the stream has gone down. Is this possible to do in a batch script?

🗪

2:25:26longboolean We should have something built into the build script that won't let you compile if the stream has gone down. Is this possible to do in a batch script?

🗪

2:27:17Update the TODO list

2:28:01mtsmox Is it an option to never clear to zero for arenas, and maybe only clear when resetting temporary memory?

🗪

2:28:01mtsmox Is it an option to never clear to zero for arenas, and maybe only clear when resetting temporary memory?

🗪

2:28:01mtsmox Is it an option to never clear to zero for arenas, and maybe only clear when resetting temporary memory?

🗪

2:30:30ctray.cpp: Investigate why the overlay disappeared

2:33:17AsafGartner Is there a still a benefit to using the sim region? Since chunks are room-sized, and simulation is room-based, why not use the chunks directly?

🗪

2:33:17AsafGartner Is there a still a benefit to using the sim region? Since chunks are room-sized, and simulation is room-based, why not use the chunks directly?

🗪

2:33:17AsafGartner Is there a still a benefit to using the sim region? Since chunks are room-sized, and simulation is room-based, why not use the chunks directly?

🗪

2:35:52Wrap it up

🗩

2:35:52Wrap it up

🗩

2:35:52Wrap it up

🗩

2:36:33Anticipate HandmadeCon 2016

🗩

2:36:33Anticipate HandmadeCon 2016

🗩

2:36:33Anticipate HandmadeCon 2016

🗩

Keyboard Navigation

Global Keys

Menu toggling

In-Menu and Index Controls

Quotes and References Menus and Index

Quotes, References and Credits Menus

Filter Menu

Filter and Link Menus

Credits Menu