Commit Graph

130 Commits

Author SHA1 Message Date
bruvzg ea1848ce0a
Use `constexpr` in the conditions with template parameters and `sizeof`s to suppress C4127 warnings. 2022-09-29 10:38:21 +03:00
Rémi Verschelde e5857bd6c7 Merge pull request from akien-mga/msvc-warnings-c4701-c4703
Fix MSVC warnings C4701 and C4703: Potentially uninitialized variable used
2022-09-28 20:47:50 +02:00
Rémi Verschelde 85fe6ecc32 Fix MSVC warnings C4701 and C4703: Potentially uninitialized variable used 2022-09-28 17:05:34 +02:00
Rémi Verschelde 0e53dd642c Fix MSVC warning C4706: assignment within conditional expression
Part of .
2022-09-28 16:05:07 +02:00
Rémi Verschelde 49fcf4ffad Style: Cleanup header guards for consistency
Fix file names for {Static,Lightmap}RaycasterEmbree.
2022-09-26 13:51:17 +02:00
Juan Linietsky f999f52f0a Add a Framebuffer cache
Adds a FramebufferCache singletion that operates the same way as UniformSetCache.

Allows creating framebuffers on the fly (and keep them cached if re-requested) such as:

```C++
RID fb = FramebufferCache::get_singleton()->get_cache(texture1,texture2);
```
2022-08-05 13:37:29 +02:00
Rémi Verschelde 57aac04480
Merge pull request from Faless/fix/4.x_warnings 2022-08-04 15:16:00 +02:00
Fabio Alessandrelli 6f02183f8c [Core] Use std type traits to check operations triviality. 2022-08-04 14:05:17 +02:00
Rémi Verschelde c717d5c64b Arrays: Zero new items of trivial types on resize() (bindings only)
This is not enabled by default in the core version for performance reasons,
as Vector/CowData are used in critical code paths where not zero'ing memory
which is going to be set later on can be important.

But for bindings / the scripting API, we make zero the new items by default
(which already happened for built types like Vector3, etc., but not for
trivial types like int, float).

Fixes .

Co-authored-by: David Hoppenbrouwers <david@salt-inc.org>
2022-08-04 13:35:37 +02:00
Rémi Verschelde 33258d850c
Merge pull request from lawnjelly/variant_bucket_pools
Variant memory pools
2022-08-02 15:54:18 +02:00
Juan Linietsky c7255388e1 Remove ThreadWorkPool, replace by WorkerThreadPool
The former needs to be allocated once per usage. The later is shared for all threads, which is more efficient.
It can also be better debugged.
2022-07-25 15:39:50 +02:00
Rémi Verschelde 90019676b0 Code quality: Fix header guards consistency
Adds `header_guards.sh` bash script, used in CI to validate future
changes. Can be run locally to fix invalid header guards.
2022-07-25 11:17:40 +02:00
reduz 455c06ecd4 Implement Vector4, Vector4i, Projection
Implement built-in classes Vector4, Vector4i and Projection.

* Two versions of Vector4 (float and integer).
* A Projection class, which is a 4x4 matrix specialized in projection types.

These types have been requested for a long time, but given they were very corner case they were not added before.
Because in Godot 4, reimplementing parts of the rendering engine is now possible, access to these types (heavily used by the rendering code) becomes a necessity.

**Q**: Why Projection and not Matrix4?
**A**: Godot does not use Matrix2, Matrix3, Matrix4x3, etc. naming convention because, within the engine, these types always have a *purpose*. As such, Godot names them: Transform2D, Transform3D or Basis. In this case, this 4x4 matrix is _always_ used as a _Projection_, hence the naming.
2022-07-23 14:00:01 +02:00
reduz 67a260d63f Implement a Worker ThreadPool
This PR implements a worked thread pool. It uses a fixed amount of threads in a pool and allows scheduling tasks
that can be run on threads (and then waited for). It satisfies the following use cases:

* HTML5 thread count is fixed (and similar restrictions are known in consoles) so we need to reuse threads.
* Thread spawning is slow in general, so reusing threads is faster anyway.
* This implementation supports recursive waiting for tasks, making it less prone to deadlocks if threads from the pool also run tasks.

After this is approved and merged, subsequent PRs will be needed to replace the ThreadWorkPool usage by this class.
2022-07-22 11:46:48 +02:00
Pedro J. Estébanez 02a584d8e6 Use the right memory ordering in SafeNumeric operations 2022-07-19 10:04:59 +02:00
bruvzg 344ba0ffaf
Refactor Font configuration and import UI, and Font resources. 2022-07-06 14:12:36 +03:00
lawnjelly b221eab426 Variant memory pools
Memory pools via PagedAllocator for Transform2D, Transform3D, Basis and AABB.
2022-07-04 12:01:46 +01:00
bruvzg 329923c6ac
Use custom key structs, instead of raw hashes for the Label3D and TextMesh, to avoid potential hash collisions. 2022-07-04 09:47:49 +03:00
Rémi Verschelde 88192269a8
Merge pull request from lyuma/packedbytearray
Prevent out-of-bounds write in array conversion; avoid logspam on empty arrays.
2022-07-01 09:03:20 +02:00
Lyuma 33fd7c63e1 Prevent out-of-bounds write in array conversion; avoid logspam on empty arrays. 2022-06-30 18:04:33 -07:00
Pedro J. Estébanez a82352c7e3 Avoid manual memory management of certain arrays in Vulkan RD 2022-06-28 10:01:46 +02:00
Hendrik Brucker fddafed919 Optimize HashMap/HashSet using fastmod 2022-06-23 18:08:52 +02:00
reduz 141c375581 Clean up Hash Functions
Clean up and do fixes to hash functions and newly introduced murmur3 hashes in 
* Clean up usage of murmur3
* Fixed usages of binary murmur3 on floats (this is invalid)
* Changed DJB2 to use xor (which seems to be better)
2022-06-20 12:54:19 +02:00
Hendrik Brucker 8c61470fa9 Hash function improvements 2022-06-15 00:32:10 +02:00
Nathan Franke 77c8f271e7
use ERR_FAIL_INDEX when preferred 2022-05-25 13:36:45 -05:00
reduz 45af29da80 Add a new HashSet template
* Intended to replace RBSet in most cases.
* Optimized for iteration speed
2022-05-20 22:40:38 +02:00
reduz 746dddc067 Replace most uses of Map by HashMap
* Map is unnecessary and inefficient in almost every case.
* Replaced by the new HashMap.
* Renamed Map to RBMap and Set to RBSet for cases that still make sense
  (order matters) but use is discouraged.

There were very few cases where replacing by HashMap was undesired because
keeping the key order was intended.
I tried to keep those (as RBMap) as much as possible, but might have missed
some. Review appreciated!
2022-05-16 10:37:48 +02:00
reduz 8b7c7f5a75 Add a new HashMap implementation
Adds a new, cleaned up, HashMap implementation.

* Uses Robin Hood Hashing (https://en.wikipedia.org/wiki/Hash_table#Robin_Hood_hashing).
* Keeps elements in a double linked list for simpler, ordered, iteration.
* Allows keeping iterators for later use in removal (Unlike Map<>, it does not do much
  for performance vs keeping the key, but helps replace old code).
* Uses a more modern C++ iterator API, deprecates the old one.
* Supports custom allocator (in case there is a wish to use a paged one).

This class aims to unify all the associative template usage and replace it by this one:
* Map<> (whereas key order does not matter, which is 99% of cases)
* HashMap<>
* OrderedHashMap<>
* OAHashMap<>
2022-05-12 11:21:29 +02:00
Haoyu Qiu 380a53f02f Add search methods for packed arrays
* count()
* find()
* rfind()
2022-05-07 20:16:11 +08:00
AndreaCatania 4066349488 Add mutable OAHashMap::lookup_ptr function to fix mutability. 2022-04-22 09:20:15 +02:00
Rémi Verschelde c3cf4d4974
Merge pull request from Pineapple/cowdata-get-data
Remove get_data() from CowData
2022-04-12 14:27:48 +02:00
Bartłomiej T. Listwon 6dfcfecd3d Remove get_data() from CowData 2022-04-09 21:25:31 +02:00
Mark Riedesel 4f3769fd75 add SafeList destructor which calls maybe_cleanup() to prevent mem leak 2022-04-08 09:50:49 -04:00
bruvzg f851c4aa33
Fix some issues found by cppcheck. 2022-04-06 14:34:37 +03:00
Rémi Verschelde f8ab79e68a Zero initialize all pointer class and struct members
This prevents the pitfall of UB when checking if they have been
assigned something valid by comparing to nullptr.
2022-04-04 19:49:50 +02:00
mashumafi 9c2bfeb2fb Const Ref Callable for custom sort/search 2022-03-27 22:10:36 -04:00
kobewi 39d429e497 Change some math macros to constexpr
Changes `MAX`, `MIN`, `ABS`, `CLAMP` and `SIGN`.
2022-03-09 16:24:32 +01:00
reduz b0ca03b0a2 Add a UniformSet cache
* Changed syntax usage for RD::Uniform to create faster with a single RID
* Converted render pass setup to use this in clustered renderer to test.

This is the first step into creating a proper uniform set cache system to simplify large parts of the codebase.
2022-03-06 13:03:33 +01:00
Rémi Verschelde 0d1e3893d9
Merge pull request from lawnjelly/bvh4_templated_checks
[4.x] BVH - Sync BVH with 3.x
2022-03-04 23:29:38 +01:00
Haoyu Qiu 3057b19daa Make VMap::find_nearest return -1 when empty 2022-02-16 16:12:30 +08:00
reduz 74adf0bf2e Remove RID_Owner.get_rid_by_index
* Implementing this function efficiently is not really possible.
* Replaced by an option to get all RIDs into a buffer for performance.
2022-02-05 11:59:34 +01:00
lawnjelly f8eaab5b47 BVH - Sync BVH with 3.x
Templated mask checks and generic NUM_TREES
Fix leaking leaves
2022-02-04 16:51:21 +00:00
Anilforextra fc27636999 Vectors: Use clear() and has().
Use clear() instead of resize(0).

Use has() instead of "find(p_val) != -1".
2022-02-02 00:11:09 +05:45
Rémi Verschelde 9912492e93
Merge pull request from akien-mga/array-slice-nicer-bound-checks 2022-01-18 13:22:35 +01:00
Rémi Verschelde 585231a172
Merge pull request from akien-mga/remove-author-docstrings 2022-01-12 15:24:17 +01:00
Rémi Verschelde c6cefb1b79
`Array`: Relax `slice` bound checks to properly handle negative indices
The same is done for `Vector` (and thus `Packed*Array`).

`begin` and `end` can now take any value and will be clamped to
`[-size(), size()]`. Negative values are a shorthand for indexing the array
from the last element upward.

`end` is given a default `INT_MAX` value (which will be clamped to `size()`)
so that the `end` parameter can be omitted to go from `begin` to the max size
of the array.

This makes `slice` works similarly to numpy's and JavaScript's.
2022-01-10 22:42:03 +01:00
Haoyu Qiu af67e4c291 Fix crash on importing FBX file 2022-01-10 18:24:31 +08:00
Haoyu Qiu c0d3bdc0ca Add list initialization support for Vector & LocalVector 2022-01-05 20:42:09 +08:00
Rémi Verschelde ba2bdc478b
Style: Remove inconsistently used `@author` docstrings
Each file in Godot has had multiple contributors who co-authored it over the
years, and the information of who was the original person to create that file
is not very relevant, especially when used so inconsistently.

`git blame` is a much better way to know who initially authored or later
modified a given chunk of code, and most IDEs now have good integration to
show this information.
2022-01-04 20:42:50 +01:00
Rémi Verschelde fe52458154
Update copyright statements to 2022
Happy new year to the wonderful Godot community!
2022-01-03 21:27:34 +01:00