Commit Graph

5395 Commits (572f4a8509965148e3dc83e8dd5dd43c2adf5cd8)

Author SHA1 Message Date
Sylvain Becker 7d47f526a7 SDL_LockTextureToSurface: robustness of locked region compared to texture size 2019-10-01 09:26:30 +02:00
Ozkan Sezer a664e95d65 dynapi: move new SDL_LockTextureToSurface addition to the end of file. 2019-10-01 00:05:50 +03:00
Sylvain Becker 1ae61f1009 Added a helper function SDL_LockTextureToSurface()
Similar to SDL_LockTexture(), except the locked area is exposed as a SDL surface.
2019-09-30 20:58:44 +02:00
Alex Szpakowski f46ac1e9f7 macOS: Fix SDL_metal.h not being copied to the framework's Headers folder when SDL.framework is built using Xcode. 2019-09-27 20:15:42 -03:00
Sam Lantinga 191fe25b1a Fixed rtkit feature guard 2019-09-27 15:21:31 -07:00
Sam Lantinga 84dc44e24d Fix the RealtimeKit dbus include guards
The SDL_USE_LIBDBUS define is set inside SDL_debug.h, therefore the
circular dependency made it impossible for this feature to be enabled.

Instead, guard SDL_dbus.h based on the autoconf variable HAVE_DBUS_DBUS_H

Additionally, fix one of the rtkit comments. CAP_SYS_NICE isn't required
to achieve high priority. But there is some scheduler config that rtkit
needs the app to setup.
2019-09-27 15:21:27 -07:00
Ryan C. Gordon 4001e6b351 stdlib: Patched to compile. 2019-09-26 13:44:49 -04:00
Ryan C. Gordon 987aa3113c stdlib: Try to coerce VS2019 to not replace some loops with memset() calls.
Fixes (?) Bugzilla #4759.
2019-09-26 12:55:05 -04:00
Ryan C. Gordon aef1ed4ac6 audio: Set (something close to) the correct silence value for U16 audio.
Partially fixes Bugzilla #4805.
2019-09-25 15:40:27 -04:00
Ryan C. Gordon 693755f0b2 coreaudio: Apple doesn't support U16 data, so convert in that case. 2019-09-25 15:07:07 -04:00
Brandon Schaefer ee6d504ae9 offscreen: Define missing define on a older EGL for an EXT function which it wont have 2019-09-24 17:49:53 -04:00
Brandon Schaefer 8a41948e7f offscreen: Avoid using EGLDeviceEXT as it causes issues with older platforms, which is just a void* 2019-09-24 17:27:51 -04:00
Brandon Schaefer 68985371a0 offscreen: Add new video driver backend Offscreen
The Offscreen video driver is intended to be used for headless rendering
  as well as allows for multiple GPUs to be used for headless rendering

Currently only supports EGL (OpenGL / ES) or Framebuffers
Adds a hint to specifiy which EGL device to use: SDL_HINT_EGL_DEVICE
Adds testoffscreen.c which can be used to test the backend out
Disabled by default for now
2019-09-24 16:36:48 -04:00
David Ludwig 3d55a51482 CMake: tabs to spaces in new parts of README-cmake.md 2019-09-23 18:30:22 -04:00
David Ludwig 00bc7f6b96 CMake: document iOS/tvOS support in README-cmake.md 2019-09-23 18:27:14 -04:00
David Ludwig 6398abe588 CMake: bug-fix for tvOS support
tvOS Device support wasn't working, at least not with the current-latest tvOS release (13.0), with CMake failing during its configuration stage.
2019-09-23 18:24:03 -04:00
David Ludwig 4e518f98a8 CMake: add version strings to Apple Info.plist files
This fills in the CFBundleVersion and CFBundleShortVersionString fields,
if and when SDL's test-apps are built via CMake.  This is needed to install
the .app bundles on iOS 13+ (using 'xcrun simctl install booted path/to/testsuchandsuch.app')
2019-09-23 17:48:14 -04:00
David Ludwig ec65a34b8e CMake: tvOS support/fixes
To use, set the following CMake variables when running CMake's configuration stage:
- CMAKE_SYSTEM_NAME=tvOS
- CMAKE_OSX_SYSROOT=<SDK>  (examples: appletvos, appletvsimulator, appletvos12.4, /full/path/to/AppleTVOS.sdk, etc.)
- CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;x86_64")
2019-08-27 12:30:20 -04:00
David Ludwig b13c951cca CMake: iOS support added
When using a recent version of CMake (3.14+), this should make it possible to:
- build SDL for iOS, both static and dynamic
- build SDL test apps (as iOS .app bundles)
- generate a working SDL_config.h for iOS (using SDL_config.h.cmake as a basis)

To use, set the following CMake variables when running CMake's configuration stage:
- CMAKE_SYSTEM_NAME=iOS
- CMAKE_OSX_SYSROOT=<SDK>  (examples: iphoneos, iphonesimulator, iphoneos12.4, /full/path/to/iPhoneOS.sdk, etc.)
- CMAKE_OSX_ARCHITECTURES=<semicolon-separated list of CPU architectures> (example: "arm64;armv7s")

Examples:
- for Simulator, using the latest, installed SDK:
    cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64

- for Device, using the latest, installed SDK, 64-bit only
    cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES=arm64

- for Device, using the latest, installed SDK, mixed 32/64 bit
    cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos -DCMAKE_OSX_ARCHITECTURES="arm64;armv7s"

- for Device, using a specific SDK revision (iOS 12.4, in this example):
    cmake path/to/SDL -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphoneos12.4 -DCMAKE_OSX_ARCHITECTURES=arm64

- for Simulator, using the latest, installed SDK, and building SDL test apps (as .app bundles):
    cmake path/to/SDL -DSDL_TEST=1 -DCMAKE_SYSTEM_NAME=iOS -DCMAKE_OSX_SYSROOT=iphonesimulator -DCMAKE_OSX_ARCHITECTURES=x86_64
2019-08-27 11:07:43 -04:00
Jinke Fan abd1dd6617 Add Hygon Dhyana processor support
Background:
    Chengdu Haiguang IC Design Co., Ltd (Hygon) is a Joint Venture
    between AMD and Haiguang Information Technology Co.,Ltd., aims at
    providing high performance x86 processor for China server market.
    Its first generation processor codename is Dhyana, which
    originates from AMD technology and shares most of the
    architecture with AMD's family 17h, but with different CPU Vendor
    ID("HygonGenuine")/Family series number(Family 18h).

Related Hygon kernel patch can be found on:
http://lkml.kernel.org/r/5ce86123a7b9dad925ac583d88d2f921040e859b.1538583282.git.puwen@hygon.cn

Best regards.
2019-05-15 19:54:36 +08:00
Ozkan Sezer a74d33b796 SDL_messagebox.h: remove comma at end of enumerator list 2019-09-22 21:41:20 +03:00
Ozkan Sezer 213de48887 update version number in Makefile.os2 too 2019-09-22 20:47:00 +03:00
Sam Lantinga 3fe2d8368c Updated SDL development builds to version 2.0.11 2019-09-22 10:37:16 -07:00
Sam Lantinga cbde0ffa3d The PS4 is ignoring the volume values in the report, so we don't need to fill them in. 2019-09-19 16:50:49 -07:00
Sam Lantinga 3efea5ea28 Don't have Windows headers define min/max, in case they're defined by application code 2019-09-11 15:08:37 -07:00
Sylvain Becker 79e388bfd8 Fixed bug 4798 - PNG w/transparency breaks in SDL 2.0.10 but works in SDL 2.0.9 2019-09-10 17:12:34 +02:00
Ozkan Sezer 32bb8b4b40 test: replace some exit()s with returns. 2019-09-10 10:03:20 +03:00
Sam Lantinga 791df27a30 Fixed compiler warning on Android 2019-09-09 13:50:46 -07:00
Sam Lantinga 715e070d29 SDL_blit_N.c: Correct vec_perm() application on little-endian 64-bit PowerPC
The LE transformation for vec_perm has an implicit assumption that the
permutation is being used to reorder vector elements (in this case 4-byte
integer word elements), not to reorder bytes within those elements.  Although
this is legal behavior, it is not anticipated by the transformation performed
by the compilers.

This causes pygame-1.9.1 test failure on PPC64LE because blitted pixmaps are
corrupted there due to how SDL uses vec_perm().

From RedHat / Fedora: https://bugzilla.redhat.com/show_bug.cgi?id=1392465
Original patch was provided by: Menanteau Guy <menantea@linux.vnet.ibm.com>
2019-09-06 08:50:19 -07:00
Conn O'Griofa 2d37d29183 KMSDRM_GLES_SwapWindow: fix non-vsync case
If KMSDRM_drmModeSetCrtc is called when the swap interval is
set to 0, the driver behaves as though vertical sync is engaged by
limiting framerate to the refresh rate, but performance is much worse
than with vertical sync enabled.

Resolve this issue by ensuring that the Crtc is only set up once,
and KMSDRM_drmModePageFlip is called, albeit without any followup
queueing or waiting for flips.
2019-09-06 08:44:46 -07:00
Sam Lantinga aaec90e5c5 Fixed bug 4789 - Linux accelerometers no longer available as joysticks
Daniel Drake

A long time ago, it was possible to play neverball on Linux using the accelerometer found in HP laptops.

The kernel exposes the accelerometer as a joystick (/dev/input/jsX) as well as an evdev device (/dev/input/eventX). I guess it worked fine when SDL was using the js interface, but then stopped working here: http://hg.libsdl.org/SDL/rev/fdaeea9e7567

Looking at current code which uses udev to discover joysticks, it looks for the udev tag ID_INPUT_JOYSTICK.

However udev's internal input_id logic specifically tags accelerometers as ID_INPUT_ACCELEROMETER and nothing else.

This looks like a good fit for SDL_HINT_ACCELEROMETER_AS_JOYSTICK.
2019-09-06 08:42:54 -07:00
Ozkan Sezer 8a394209c4 SDL_windowsevents.c: remove isVistaOrNewer (not used since 8cb1dc50bb28) 2019-09-05 20:47:20 +03:00
Ozkan Sezer 51df932efb fix permissions 2019-09-05 13:21:02 +03:00
Sylvain Becker bf2f4703f2 SDL_windowsmessagebox.c: remove unused variable 2019-09-05 10:49:53 +02:00
Sylvain Becker 1b5e3c19a2 SDL_bmp.c: remove unused variable warnings 2019-09-05 10:08:47 +02:00
Sam Lantinga e5580e18ba x11: add a hint to force the VisualID used when creating a window. 2019-09-04 09:27:58 -07:00
Ryan C. Gordon f49c07b5c4 stdinc: On macOS and iOS, use memset_pattern4() for SDL_memset4().
Fixes Bugzilla #4724.
2019-09-04 00:39:47 -04:00
Sam Lantinga 892c8d5058 Fixed bug 4536 - Heap-Buffer Overflow in SDL_GetRGB pertaining to SDL_pixels.c
Ozkan Sezer

As for the issue: This bmp reports bpp=0, therefore SDL_CalculatePitch()
returns pitch==0, which is then fed to SDL_malloc() (which is malloc())
and malloc(0) returns _something_ which is not NULL but not someting
that we expect..  Then testsprite.c:LoadSprite() accesses the pixels
as *(Uint8*)pixels which valrind reports as:

==15533== Invalid read of size 1
==15533==    at 0x8048C08: LoadSprite (testsprite.c:45)
==15533==    by 0x80492FC: main (testsprite.c:224)
==15533==  Address 0x449e588 is 0 bytes after a block of size 0 alloc'd
==15533==    at 0x40072B2: malloc (vg_replace_malloc.c:270)
==15533==    by 0x4045719: SDL_CreateRGBSurface (SDL_surface.c:126)
==15533==    by 0x40403C1: SDL_LoadBMP_RW (SDL_bmp.c:237)
==15533==    by 0x8048BB2: LoadSprite (testsprite.c:36)
==15533==    by 0x80492FC: main (testsprite.c:224)

Besides, valrind also reports this:
==15533== Conditional jump or move depends on uninitialised value(s)
==15533==    at 0x40403F3: SDL_LoadBMP_RW (SDL_bmp.c:247)
==15533==    by 0x8048BB2: LoadSprite (testsprite.c:36)
==15533==    by 0x80492FC: main (testsprite.c:224)


Easy/quick solution would be early-rejecting a bmp with 0 bpp from SDL_bmp.c:SDL_LoadBMP_RW()
2019-09-03 11:55:20 -07:00
Ozkan Sezer b21338eb54 SDL_bmp.c: restore most of the original formatting. 2019-09-02 12:35:00 +03:00
Ryan C. Gordon 847bd8d983 direct3d: Be more aggressive about resetting state when textures go away.
Fixes Bugzilla #4768.
2019-09-02 00:11:58 -04:00
Ryan C. Gordon a3804ba1da d3d11: Fixed VB state, avoiding unnecessary recreation (Thanks, Alex!).
Fixes Bugzilla #4779.
2019-09-01 22:41:44 -04:00
Sylvain Becker 6c295129ba LoadBMP: fix some warnings 2019-08-31 22:58:11 +02:00
Sylvain Becker 830979c555 LoadBMP: use code from SDL_image which allows loading compressed BMP files 2019-08-31 22:52:15 +02:00
Sam Lantinga cc64b369fb Allow mouse messages from Wacom pens, e.g. right click pen buttons, etc. 2019-08-30 15:32:15 -07:00
Sam Lantinga afb9ff9507 Fixed bug where the Steam overlay would generate an event and stop input processing for the frame. 2019-08-30 08:03:19 -07:00
Ozkan Sezer 9c8e403f6b use 'U' suffix on constants instead of (unsigned int) cast. 2019-08-30 11:35:20 +03:00
Sylvain Becker d52080c0ab Android: minor warning 2019-08-30 09:00:06 +02:00
Sylvain Becker 70dc8d1648 Android: fix corresponding warnings 2019-08-30 08:55:20 +02:00
Sylvain Becker 6794ec6670 Android: add more warning flags 2019-08-30 08:43:13 +02:00
Sylvain Becker 0a9c74aa9a Fixed bug 3918 - HIDAPI, CMake support for android project 2019-08-27 11:38:43 +02:00