Commit Graph

2245 Commits

Author SHA1 Message Date
clayjohn 3db61ce91c Avoid material rebinds when using skeletons 2020-04-07 10:34:35 -07:00
Rémi Verschelde 597ec5dbf8
Merge pull request #37586 from qarmin/vulkan_allocator_leak
Fix leak with Vulkan allocator
2020-04-07 09:16:02 +02:00
Rémi Verschelde befbdda7f2
Merge pull request #37635 from qarmin/leak_vertex
Fixed leak with vertex_formats
2020-04-06 21:32:25 +02:00
qarmin f4301b7259 Fixed leak with vertex_formats 2020-04-06 18:07:31 +02:00
Rémi Verschelde 61342e5831 Vulkan: Use VK_NULL_HANDLE for handles, fixes 32-bit builds
Some Vulkan types are defined as "non dispatchable handles" and use a
different typedef on 32-bit and 64-bit systems (struct pointer on
64-bit, `uint64_t` otherwise).

0e78ffd1dc/include/vulkan/vulkan_core.h (L59-L65)

Contrarily to `NULL`, `nullptr` can't be converted to `uint64_t` so
build was now failing on 32-bit after converting the codebase from
using `NULL` to `nullptr`.

Fixes #37620.
2020-04-06 10:21:32 +02:00
qarmin a0832e6f5e Fix leak with Vulkan allocator 2020-04-04 20:16:48 +02:00
lupoDharkael 95a1400a2a Replace NULL with nullptr 2020-04-02 13:38:00 +02:00
Pedro J. Estébanez 741f8d9bec Fix res:// trimmed to s:// on Windows 2020-04-01 10:30:22 +02:00
Rémi Verschelde e5033d3952 doc: Sync classref with DisplayServer/Window changes 2020-03-31 11:56:58 +02:00
Rémi Verschelde cd4e46ee65 SCons: Format buildsystem files with psf/black
Configured for a max line length of 120 characters.

psf/black is very opinionated and purposely doesn't leave much room for
configuration. The output is mostly OK so that should be fine for us,
but some things worth noting:

- Manually wrapped strings will be reflowed, so by using a line length
  of 120 for the sake of preserving readability for our long command
  calls, it also means that some manually wrapped strings are back on
  the same line and should be manually merged again.

- Code generators using string concatenation extensively look awful,
  since black puts each operand on a single line. We need to refactor
  these generators to use more pythonic string formatting, for which
  many options are available (`%`, `format` or f-strings).

- CI checks and a pre-commit hook will be added to ensure that future
  buildsystem changes are well-formatted.
2020-03-30 09:05:53 +02:00
Juan Linietsky a6f3bc7c69 Renaming of servers for coherency.
VisualServer -> RenderingServer
PhysicsServer -> PhysicsServer3D
Physics2DServer -> PhysicsServer2D
NavigationServer -> NavigationServer3D
Navigation2DServer -> NavigationServer2D

Also renamed corresponding files.
2020-03-27 15:21:27 -03:00
Juan Linietsky 9e08742de8 Added a Window node, and made it the scene root.
Still a lot of work to do.
2020-03-26 15:49:38 +01:00
Juan Linietsky f8a79a97c7 Effective DisplayServer separation, rename X11 -> LinuxBSD 2020-03-26 15:49:34 +01:00
Rémi Verschelde cbbe0743a9
Merge pull request #37219 from RajatGoswami/missing-include-guards
Adding missing include guards to header files identified by LGTM
2020-03-23 11:17:24 +01:00
Rajat Goswami 2ecf928ae3 Adding missing include guards to header files identified by LGTM.
This addresses the issue godotengine/godot#37143
2020-03-23 04:52:36 -04:00
Rémi Verschelde ed9a0d0484
Merge pull request #37179 from clayjohn/VULKAN-sky-shader
Implement Sky Shaders
2020-03-22 20:03:35 +01:00
clayjohn 61a74739ca Working sky shader implementation 2020-03-21 20:43:44 -07:00
Aaron Franke 7dbe8b65ae
Make file formatting comply with POSIX and Unix standards
UTF-8, LF, no BOM, and newlines at the end of files
2020-03-21 17:41:03 -04:00
Bastiaan Olij c3fee7ba6c Add shader based background mode 2020-03-19 18:30:39 -07:00
Rémi Verschelde cb282c6ef0 Style: Set clang-format Standard to Cpp11
For us, it practically only changes the fact that `A<A<int>>` is now
used instead of the C++03 compatible `A<A<int> >`.

Note: clang-format 10+ changed the `Standard` arguments to fully
specified `c++11`, `c++14`, etc. versions, but we can't use `c++17`
now if we want to preserve compatibility with clang-format 8 and 9.
`Cpp11` is still supported as deprecated alias for `Latest`.
2020-03-17 07:36:24 +01:00
Rémi Verschelde 05d9d1c0e7
Merge pull request #36905 from Faless/js/restore_and_ci
Resurrect HTML5 platform, add it to CI (no rendering yet)
2020-03-11 07:31:16 +01:00
Rémi Verschelde f70b5fa30c Revert "Enhanced physical device selection to use device type and available memory in relation to issue #35397"
This reverts commit 4f3006e5ac.
2020-03-10 08:58:34 +01:00
Haoyu Qiu 46230d253b Initializes VulkanContext::queue_props to NULL 2020-03-10 14:36:39 +08:00
Rémi Verschelde 478337c412
Merge pull request #36751 from Faless/debugger/threads_and_profilers
ScriptDebugger refactor, threading, profilers.
2020-03-09 19:08:07 +01:00
PouleyKetchoupp d99bea20dc More explicit error messages when vulkan calls return errors 2020-03-09 13:25:54 +01:00
Fabio Alessandrelli b8ddaf9c33 Refactor ScriptDebugger.
EngineDebugger is the new interface to access the debugger.
It tries to be as agnostic as possible on the data that various
subsystems can expose.

It allows 2 types of interactions:

- Profilers:
  A subsystem can register a profiler, assigning it a unique name.
  That name can be used to activate the profiler or add data to it.
  The registered profiler can be composed of up to 3 functions:
    - Toggle: called when the profiler is activated/deactivated.
    - Add: called whenever data is added to the debugger
      (via `EngineDebugger::profiler_add_frame_data`)
    - Tick: called every frame (during idle), receives frame times.

- Captures: (Only relevant in remote debugger for now)
  A subsystem can register a capture, assigning it a unique name.
  When receiving a message, the remote debugger will check if it starts
  with `[prefix]:` and call the associated capture with name `prefix`.

Port MultiplayerAPI, Servers, Scripts, Visual, Performance to the new
profiler system.

Port SceneDebugger and RemoteDebugger to the new capture system.
The LocalDebugger also uses the new profiler system for scripts
profiling.
2020-03-08 12:36:39 +01:00
Fabio Alessandrelli aa8c825855 OS_Javascript temporarly uses dummy rasterizer. 2020-03-08 03:35:13 +01:00
Fabio Alessandrelli 7b22ca339f Restore dummy rasterizer
Not all functionalities are back, but it compiles and runs minimal
project.
2020-03-08 01:55:27 +01:00
Rémi Verschelde 42595085a5
Merge pull request #36752 from RandomShaper/rework_semaphore
Drop old semaphore implementation
2020-03-05 16:33:45 +01:00
Rémi Verschelde 398f714dfc
Merge pull request #36792 from akien-mga/vulkan-better-debug-prints
Vulkan: Use Godot print macros matching MessageSeverity flag
2020-03-04 20:31:02 +01:00
Rémi Verschelde 3625b057af Vulkan: Use Godot print macros matching MessageSeverity flag
Otherwise any verbose/info/warning debug message from Vulkan would
raise an error, confusing users about the severity of the message.

Cf. #36185, #36790.
2020-03-04 15:02:49 +01:00
Rémi Verschelde a0e33e17fb
Merge pull request #36733 from qarmin/static_analyzer_fixes
Fixes bugs found by Sonarcloud and Coverity
2020-03-04 13:41:38 +01:00
Rémi Verschelde f83f1d7c9b
Merge pull request #36069 from RandomShaper/imvu/improve_drives_ux
Improve UX of drive letters
2020-03-04 13:19:55 +01:00
Rémi Verschelde 2856740e6b
Merge pull request #36385 from kuruk-mm/some_static_asserts
Change when we can ERR_FAIL_COND and ERR_FAIL_CONV_V to static_assert
2020-03-04 13:18:48 +01:00
Rémi Verschelde 98f88a4102
Merge pull request #36585 from disks86/master
Enhanced physical device selection
2020-03-04 13:00:52 +01:00
Mateo Dev .59 c933eb7acc Change when we can ERR_FAIL_COND and ERR_FAIL_CONV_V to static_assert 2020-03-04 04:44:47 -03:00
Mateo Dev .59 8a88637705 os: execute parse the command output from utf8 2020-03-03 21:58:50 -03:00
Pedro J. Estébanez 9a3a2b03b8 Drop old semaphore implementation
- Removed platform-specific implementations.
- Now all semaphores are in-object, unless they need to be conditionally created.
- Similarly to `Mutex`, provided a dummy implementation for when `NO_THREADS` is defined.
- Similarly to `Mutex`, methods are made `const` for easy use in such contexts.
- Language bindings updated: `wait()` and `post()` are now `void`.
- Language bindings updated: `try_wait()` added.

Bonus:
- Rewritten the `#ifdef` in `mutex.h` to meet the code style.
2020-03-03 13:20:42 +01:00
Pedro J. Estébanez aee586553a Improve UX of drive letters
Namely, move the drive dropdown to just the left of the path text box and don't include the former
in the latter.

This improves the UX on Windows.

In the UNIX case, since its concept of drives is (ab)used to provide shortcuts to useful paths, its
dropdown is kept at the original location.
2020-03-03 10:38:34 +01:00
qarmin 1f209bfc41 Fixes bugs found by Sonarcloud and Coverity 2020-03-02 19:17:20 +01:00
Haoyu Qiu 771e66b080 Adds null check of pMessageIdName 2020-02-29 18:26:38 +08:00
Christopher Joseph Dean Schaefer 4f3006e5ac Enhanced physical device selection to use device type and available memory in relation to issue #35397 2020-02-28 21:02:29 -05:00
Juan Linietsky 475e4ea67b Removed interactive loader, added proper thread loading. 2020-02-28 11:20:45 -03:00
Rémi Verschelde b7b3978684
Merge pull request #36556 from RandomShaper/rework_mutex
Reimplement `Mutex` with C++'s `<mutex>` (plus more)
2020-02-28 00:26:01 +01:00
Rémi Verschelde 3a7f1300ee
Merge pull request #36186 from Calinou/remove-vulkan-debugging-prints
Remove Vulkan debugging prints
2020-02-27 21:11:24 +01:00
Pedro J. Estébanez 18fbdbb456 Reimplement Mutex with C++'s <mutex>
Main:
- It's now implemented thanks to `<mutex>`. No more platform-specific implementations.
- `BinaryMutex` (non-recursive) is added, as an alternative for special cases.
- Doesn't need allocation/deallocation anymore. It can live in the stack and be part of other classes.
- Because of that, it's methods are now `const` and the inner mutex is `mutable` so it can be easily used in `const` contexts.
- A no-op implementation is provided if `NO_THREADS` is defined. No more need to add `#ifdef NO_THREADS` just for this.
- `MutexLock` now takes a reference. At this point the cases of null `Mutex`es are rare. If you ever need that, just don't use `MutexLock`.
- Thread-safe utilities are therefore simpler now.

Misc.:
- `ScopedMutexLock` is dropped and replaced by `MutexLock`, because they were pretty much the same.
- Every case of lock, do-something, unlock is replaced by `MutexLock` (complex cases where it's not straightfoward are kept as as explicit lock and unlock).
- `ShaderRD` contained an `std::mutex`, which has been replaced by `Mutex`.
2020-02-26 20:40:10 +01:00
Juan Linietsky 33b5c57199 Variant: Added 64-bit packed arrays, renamed Variant::REAL to FLOAT.
- Renames PackedIntArray to PackedInt32Array.
- Renames PackedFloatArray to PackedFloat32Array.
- Adds PackedInt64Array and PackedFloat64Array.
- Renames Variant::REAL to Variant::FLOAT for consistency.

Packed arrays are for storing large amount of data and creating stuff like
meshes, buffers. textures, etc. Forcing them to be 64 is a huge waste of
memory. That said, many users requested the ability to have 64 bits packed
arrays for their games, so this is just an optional added type.

For Variant, the float datatype is always 64 bits, and exposed as `float`.

We still have `real_t` which is the datatype that can change from 32 to 64
bits depending on a compile flag (not entirely working right now, but that's
the idea). It affects math related datatypes and code only.

Neither Variant nor PackedArray make use of real_t, which is only intended
for math precision, so the term is removed from there to keep only float.
2020-02-25 12:55:53 +01:00
Rémi Verschelde da3a504216
Merge pull request #36454 from akien-mga/vulkan-workaround-32-bit-lib-error
Vulkan: Work around false positive on 64-bit Linux w/ 32-bit ICDs
2020-02-22 16:28:48 +01:00
Juan Linietsky 54dfdd1cdb Correct condition wrongly converted to ERR_FAIL_COND_MSG
Fixes debugging of giprobes not working, likely other stuff
2020-02-22 11:36:26 -03:00
Rémi Verschelde e596439497 Vulkan: Work around false positive on 64-bit Linux w/ 32-bit ICDs
In the vast majority of cases, this will be a false positive error
thrown by Vulkan-Loader when a Linux system has Vulkan ICDs for both
32-bit and 64-bit. The error is of the form:
```
ERROR: [Loader Message] Code 0 : /usr/lib/libvulkan_intel.so: wrong ELF class: ELFCLASS32
ERROR: [Loader Message] Code 0 : /usr/lib/libvulkan_radeon.so: wrong ELF class: ELFCLASS32
```

The loader dlopen's the 32-bit ICDs first, raises this error, and then
happily goes on to try and use the 64-bit ICDs.

Upstream report: https://github.com/KhronosGroup/Vulkan-Loader/issues/262

Fixes #36185.
2020-02-22 15:32:20 +01:00