Many contributors (me included) did not fully understand what CCFLAGS,
CXXFLAGS and CPPFLAGS refer to exactly, and were thus not using them
in the way they are intended to be.
As per the SCons manual: https://www.scons.org/doc/HTML/scons-user/apa.html
- CCFLAGS: General options that are passed to the C and C++ compilers.
- CFLAGS: General options that are passed to the C compiler (C only;
not C++).
- CXXFLAGS: General options that are passed to the C++ compiler. By
default, this includes the value of $CCFLAGS, so that setting
$CCFLAGS affects both C and C++ compilation.
- CPPFLAGS: User-specified C preprocessor options. These will be
included in any command that uses the C preprocessor, including not
just compilation of C and C++ source files [...], but also [...]
Fortran [...] and [...] assembly language source file[s].
TL;DR: Compiler options go to CCFLAGS, unless they must be restricted
to either C (CFLAGS) or C++ (CXXFLAGS). Preprocessor defines go to
CPPFLAGS.
It seems to stay compatible with formatting done by clang-format 6.0 and 7.0,
so contributors can keep using those versions for now (they will not undo those
changes).
Godot supports many different compilers and for production releases we
have to support 3 currently: GCC8, Clang6, and MSVC2017. These compilers
all do slightly different things with -ffast-math and it is causing
issues now. See #24841, #24540, #10758, #10070. And probably other
complaints about physics differences between release and release_debug
builds.
I've done some performance comparisons on Linux x86_64. All tests are
ran 20 times.
Bunnymark: (higher is better)
(bunnies) min max stdev average
fast-math 7332 7597 71 7432
this pr 7379 7779 108 7621 (102%)
FPBench (gdscript port http://fpbench.org/) (lower is better)
(ms)
fast-math 15441 16127 192 15764
this pr 15671 16855 326 16001 (99%)
Float_add (adding floats in a tight loop) (lower is better)
(sec)
fast-math 5.49 5.78 0.07 5.65
this pr 5.65 5.90 0.06 5.76 (98%)
Float_div (dividing floats in a tight loop) (lower is better)
(sec)
fast-math 11.70 12.36 0.18 11.99
this pr 11.92 12.32 0.12 12.12 (99%)
Float_mul (multiplying floats in a tight loop) (lower is better)
(sec)
fast-math 11.72 12.17 0.12 11.93
this pr 12.01 12.62 0.17 12.26 (97%)
I have also looked at FPS numbers for tps-demo, 3d platformer, 2d
platformer, and sponza and could not find any measurable difference.
I believe that given the issues and oft-reported (physics) glitches on
release builds I believe that the couple of percent of tight-loop
floating point performance regression is well worth it.
This fixes#24540 and fixes#24841
Fixes the following GCC 5 warnings:
```
core/io/zip_io.h:128:26: warning: 'zlib_filefunc_def zipio_create_io_from_file(FileAccess**)' defined but not used [-Wunused-function]
core/script_debugger_remote.cpp:110:17: warning: 'ObjectID safe_get_instance_id(const Variant&)' defined but not used [-Wunused-function]
drivers/unix/socket_helpers.h:103:12: warning: 'int _socket_create(IP::Type&, int, int)' defined but not used [-Wunused-function]
drivers/unix/socket_helpers.h:45:15: warning: 'size_t _set_sockaddr(sockaddr_storage*, const IP_Address&, int, IP::Type)' defined but not used [-Wunused-function]
drivers/unix/socket_helpers.h:76:15: warning: 'size_t _set_listen_sockaddr(sockaddr_storage*, int, IP::Type, IP_Address)' defined but not used [-Wunused-function]
editor/editor_fonts.cpp:40:24: warning: 'Ref<BitmapFont> make_font(int, int, int, int, const int*, const Ref<Texture>&)' defined but not used [-Wunused-function]
editor/editor_themes.cpp:85:26: warning: 'Ref<StyleBoxFlat> change_border_color(Ref<StyleBoxFlat>, Color)' defined but not used [-Wunused-function]
editor/import/editor_import_collada.cpp:493:13: warning: 'void _generate_normals(const PoolVector<int>&, const PoolVector<Vector3>&, PoolVector<Vector3>&)' defined but not used [-Wunused-function]
editor/import/editor_import_collada.cpp:524:13: warning: 'void _generate_tangents_and_binormals(const PoolVector<int>&, const PoolVector<Vector3>&, const PoolVector<Vector3>&, const PoolVector<Vector3>&, PoolVector<float>&)' defined but not used [-Wunused-function]
editor/pvrtc_compress.cpp:118:13: warning: 'void _compress_etc(Image*)' defined but not used [-Wunused-function]
modules/etc/image_etc.cpp:89:13: warning: 'void _decompress_etc1(Image*)' defined but not used [-Wunused-function]
modules/etc/image_etc.cpp:93:13: warning: 'void _decompress_etc2(Image*)' defined but not used [-Wunused-function]
modules/gdscript/editor/gdscript_highlighter.cpp:46:13: warning: 'bool _is_whitespace(CharType)' defined but not used [-Wunused-function]
scene/2d/cpu_particles_2d.cpp:510:14: warning: 'float rand_from_seed_m1_p1(uint32_t&)' defined but not used [-Wunused-function]
scene/3d/cpu_particles.cpp:474:14: warning: 'float rand_from_seed_m1_p1(uint32_t&)' defined but not used [-Wunused-function]
scene/resources/default_theme/default_theme.cpp:123:20: warning: 'Ref<Shader> make_shader(const char*, const char*, const char*)' defined but not used [-Wunused-function]
scene/resources/default_theme/default_theme.cpp:130:24: warning: 'Ref<BitmapFont> make_font(int, int, int, int, const int*, const Ref<Texture>&)' defined but not used [-Wunused-function]
```
Had to split `core/io/zip_io.h` into header and .cpp file without 'static' keyword.
Not fixed yet (static definition in header used in some files but not all):
```
modules/websocket/lws_helper.h:111:13: warning: 'void _lws_make_protocols(void*, int (*)(lws*, lws_callback_reasons, void*, void*, size_t), PoolVector<String>, _LWSRef**)' defined but not used [-Wunused-function]
```
Also fixed a couple other warnings missed in previous commits.
Fixes GCC 5 warnings of the form:
core/io/http_client.cpp:288:9: warning: enumeration value 'STATUS_SSL_HANDSHAKE_ERROR' not handled in switch [-Wswitch]
core/io/marshalls.cpp:806:9: warning: enumeration value 'AABB' not handled in switch [-Wswitch]
Those can be trivial cases where adding a default fallback is the solution,
or more complex issues/hidden bugs where missed values are actually meant
to be handled.
Also turn off -Wsign-compare warnings in the future, we do not consider them important.
Fixes the following GCC 5 warnings:
```
core/node_path.cpp:279:24: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
core/oa_hash_map.h:169:21: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
core/oa_hash_map.h:314:21: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
drivers/gles2/shader_gles2.cpp:985:23: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
drivers/gles3/rasterizer_storage_gles3.cpp:1075:25: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
drivers/pulseaudio/audio_driver_pulseaudio.cpp:343:34: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
editor/editor_plugin.cpp:525:21: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
editor/editor_properties_array_dict.cpp:747:25: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
editor/plugins/spatial_editor_plugin.cpp:2078:20: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
editor/plugins/spatial_editor_plugin.cpp:4096:27: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
editor/plugins/sprite_editor_plugin.cpp💯20: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
modules/cvtt/image_compress_cvtt.cpp:122:23: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
modules/cvtt/image_compress_cvtt.cpp:134:77: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
modules/cvtt/image_compress_cvtt.cpp:339:25: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
modules/etc/image_etc.cpp:222:34: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
modules/gdnative/register_types.cpp:242:21: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
modules/gdnative/register_types.cpp:258:21: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
modules/opensimplex/simplex_noise.cpp:200:13: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
modules/opensimplex/simplex_noise.cpp:222:13: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
modules/opensimplex/simplex_noise.cpp:246:13: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
platform/android/export/export.cpp:1085:21: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
platform/android/export/export.cpp:1489:23: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
platform/android/export/export.cpp:1623:22: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
platform/iphone/export/export.cpp:206:20: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
platform/iphone/export/export.cpp:356:20: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
platform/iphone/export/export.cpp:406:20: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
platform/iphone/export/export.cpp:493:21: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
scene/3d/audio_stream_player_3d.cpp:420:23: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
scene/resources/audio_stream_sample.cpp:565:22: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
scene/resources/audio_stream_sample.cpp:571:22: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
servers/audio/audio_rb_resampler.cpp:156:36: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
```
The following warnings were not fixed, as they implied casting for no gain:
```
core/io/packet_peer.cpp:228:38: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
core/io/resource_format_binary.cpp:109:11: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
drivers/gles2/rasterizer_scene_gles2.cpp:144:57: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
drivers/unix/file_access_unix.cpp:249:46: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
scene/3d/voxel_light_baker.cpp:889:14: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
scene/3d/voxel_light_baker.cpp:1020:14: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
scene/3d/voxel_light_baker.cpp:1154:14: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
scene/3d/voxel_light_baker.cpp:2255:38: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
scene/resources/bit_mask.cpp:336:25: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
servers/audio/audio_stream.cpp:141:49: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
servers/audio/audio_stream.cpp:150:19: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
servers/audio/audio_stream.cpp:154:19: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
servers/audio_server.cpp:86:21: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
servers/audio_server.cpp:89:17: warning: comparison between signed and unsigned integer expressions [-Wsign-compare]
```
This allows more consistency in the manner we include core headers,
where previously there would be a mix of absolute, relative and
include path-dependent includes.
Using `misc/scripts/fix_headers.py` on all Godot files.
Some missing header guards were added, and the header inclusion order
was fixed in the Bullet module.
Apparently -ffast-math generates incorrect code with recent versions of
GCC and Clang. The manual page for GCC warns about this possibility.
In my tests it doesn't actually appear to be measurably slower in this
case, and this is used in a batch process so it seems safe to disable
this.
This fixes#10758 and fixes#10070
Rémi Verschelde
Hein-Pieter van Braam
Marc Gilleron
Juan Linietsky
Matthias Hoelzl
MillionOstrich
Elliott Sales de Andrade
Ignacio Etcheverry
Marcelo Fernandez
Ferenc Arn