Commit Graph

5491 Commits

Author SHA1 Message Date
Sam Lantinga
b5e9ebbafa Fixed compiler warning
warning C4018: '<' : signed/unsigned mismatch
2019-06-08 19:12:05 -07:00
Sam Lantinga
a21b5b3018 Fixed build 2019-06-08 19:09:43 -07:00
Sam Lantinga
990e166a3b Fixed bug 3894 - Fuzzing crashes for SDL_LoadWAV
Simon Hug

I had a look at this and made some additions to SDL_wave.c.

The attached patch adds many checks and error messages. For some reason I also added A-law and ?-law decoders. Forgot exactly why... but hey, they're small.

The WAVE format is seriously underspecified (at least by the documents that are publicly available on the internet) and it's a shame Microsoft never put something better out there. The language used in them is so loose at times, it's not surprising the encoders and decoders behave very differently. The Windows Media Player doesn't even support MS ADPCM correctly.

The patch also adds some hints to make the decoder more strict at the cost of compatibility with weird WAVE files.

I still think it needs a bit of cleaning up (Not happy with the MultiplySize function. Don't like the name and other SDL code may want to use something like this too.) and some duplicated code may be folded together. It does work in this state and I have thrown all kinds of WAVE files at it. The AFL files also pass with it and some even play (obviously just noise). Crafty little fuzzer.

Any critique would be welcome. I have a fork of SDL with a audio-loadwav branch over here if someone wants to use the commenting feature of Bitbucket:

https://bitbucket.org/ChliHug/SDL

I also cobbled some Lua scripts together to create WAVE test files:

https://bitbucket.org/ChliHug/gendat
2019-06-08 19:02:42 -07:00
Sam Lantinga
48ac92af54 Fixed bug 4041 - Android, SDL_Renderer OpenGLES 1 is loading GLESv2 library
Sylvain

On Android, if you set no attribute using SDL_GL_SetAttribute(), and try to create a SDL Render OpenGLES 1:

- it loads first by default GLESv2 libraries
- creates the rendere OpenGLES 1
- recreates the Window to have a context 1.1 ( https://hg.libsdl.org/SDL/file/4db4cfd59470/src/render/opengles/SDL_render_gles.c#l298 )

But it doesn't unload libraries, then reload GLESv1 lib. So the SDL_Renderer OpenGLES 1 is working with GLES 2 libs, which seems inconsistent.


If you, at first, set
SDL_GL_SetAttribute(SDL_GL_CONTEXT_PROFILE_MASK, SDL_GL_CONTEXT_PROFILE_ES);
SDL_GL_SetAttribute(SDL_GL_CONTEXT_MAJOR_VERSION, 1);
SDL_GL_SetAttribute(SDL_GL_CONTEXT_MINOR_VERSION, 1);
It will correctly load GLES v1 libraries.

Here's a small patch to reload egl libs when SDL_RecreateWindow() is called.
It fixes the issue, also the case from bug 4042

( SDL_RecreateWindow() is used by SDL_Renderer gl, gles1, gles2. )
2019-06-08 18:40:11 -07:00
Sam Lantinga
1b73d578f3 Temporary fix for bug 4254 - a _lot_ of strict aliasing warnings
Ozkan Sezer

A horde of strict aliasing violation warnings are emitted from joystick
layer, and also from a few other places. This happens with gcc-4.4.7 on
Linux CentOS 6.10.  Some other sysjoystick would possibly have the same
warnings.

Attached my full log here. Example entry:
src/joystick/SDL_joystick.c: In function 'SDL_GetJoystickGUIDInfo':
src/joystick/SDL_joystick.c:1094: warning: dereferencing pointer '({anonymous})' does break strict-aliasing rules
2019-06-08 18:32:29 -07:00
Sam Lantinga
31765242d6 Fixed bug 4294 - Audio: perform more validation on conversion request
janisozaur

There are many cases which are not able to be handled by SDL's audio conversion routines, including too low (negative) rate, too high rate (impossible to allocate).

This patch aims to report such issues early and handle others in a graceful manner. The "INT32_MAX / RESAMPLER_SAMPLES_PER_ZERO_CROSSING" value is the conservative approach in terms of what can _technically_ be supported, but its value is 4'194'303, or just shy of 4.2MHz. I highly doubt any sane person would use such rates, especially in SDL2, so I would like to drive this limit further down, but would need some assistance to do that, as doing so would have to introduce an arbitrary value. Are you OK with such approach? What would a good value be? Wikipedia (https://en.wikipedia.org/wiki/High-resolution_audio) lists 96kHz as the highest sampling rate in use, even if I quadruple it for a good measure, to 384kHz it's still an order of magnitude lower than 4MHz.
2019-06-08 18:22:18 -07:00
Sam Lantinga
3f19a6d5e8 CVE-2019-7578: Fix a buffer overread in InitIMA_ADPCM
If IMA ADPCM format chunk was too short, InitIMA_ADPCM() parsing it
could read past the end of chunk data. This patch fixes it.

CVE-2019-7578
https://bugzilla.libsdl.org/show_bug.cgi?id=4494

Signed-off-by: Petr P?sa? <ppisar@redhat.com>
2019-06-08 18:07:58 -07:00
Sam Lantinga
316ff3847b Fixed bug 4526 - replace SDL_RW* macros with functions for using in bindings
ace

I got this bug in SDL_ttf:
https://bugzilla.libsdl.org/show_bug.cgi?id=4524
Sylvain proposed solution:
SDL_RWseek(RWops, 0, RW_SEEK_SET);

And it works, but i can use it my project, because it written in C# with SDL2-CS wrapper and there not export for macroses:
#define SDL_RWsize(ctx)         (ctx)->size(ctx)
#define SDL_RWseek(ctx, offset, whence) (ctx)->seek(ctx, offset, whence)
#define SDL_RWtell(ctx)         (ctx)->seek(ctx, 0, RW_SEEK_CUR)
#define SDL_RWread(ctx, ptr, size, n)   (ctx)->read(ctx, ptr, size, n)
#define SDL_RWwrite(ctx, ptr, size, n)  (ctx)->write(ctx, ptr, size, n)
#define SDL_RWclose(ctx)        (ctx)->close(ctx)

Therefore, I suggest replacing this macros with functions so that they can be exported and used in bindings
2019-06-08 17:43:23 -07:00
Sam Lantinga
2b6473dc05 Fixed bug 4533 - Update ANGLE to load d3dcompiler_47.dll instead of d3dcompiler_46.dll
msmshazan

Update ANGLE Libraries to support d3dcompiler_47.dll since chrome does not ship with d3dcompiler_46.dll and d3dcompiler_43.dll
2019-06-08 15:10:20 -07:00
Ethan Lee
0442d19fc8 cocoa: Fix assert to use SDL_assert 2019-03-04 12:16:43 -05:00
Sam Lantinga
a73dacbf62 Backed out Ben's chinese Xbox controller patch, as the generic catch-all for Xbox controllers should handle it. 2019-06-08 14:58:49 -07:00
Benjamin Valentin
f3c6b1f51a Add mapping for Chinese-made Xbox Controller
This device is a copy of the Xbox Controller S and currently the one most sold
when shopping for a 'new' Xbox gamepad on eBay and AliExpress.
Except for the quirky USB ID id behaves just like a normal Xbox controller (when
ignoring the subpar build quality)
2019-03-17 23:47:12 +01:00
Sam Lantinga
8728ce4448 Fixed bug 4557 - SDL_SIMDAlloc and *Free should be in the public interface
Martin Gerhardy

These functions are really useful and should get exposed imo.
2019-06-08 14:54:37 -07:00
Sam Lantinga
56b7f4cf31 Fixed bug 4583 - PollAllValues appears to use an incorrect index for all axes above 0x18
Noam Preil

In src/joystick/linux/SDL_sysjoystick.c:

The ConfigJoystick function's axes detection starts with a for loop using an index i for Linux's axes names. When i gets to ABS_HAT0X, it's set to ABS_HAT3Y and a continue statement appears, to skip the hats. This makes sense, as SDL handles hats separately from axes.

However, in PollAllValues, *two* indices are used: a and b. Both start out the same, and remain so until the hats are reached. At that point, a becomes identical to the i from ConfigJoystick's loop, but b is equal to a - (ABS_HAT3Y - ABS_HAT0X), or a - 8.

While all the joystick->hwdata->abs_* structures in ConfigJoystick used i - which would here be a - as both the index and the ioctl argument, PollAllValues uses b for the structure index and a as the ioctl argument.

It would appear, however, that no joystick HAS such axes, and that the b index is entirely unnecessary.

I tested three separate joysticks, and while that was far from a complete listing, I was unable to find a joystick with an axis above 0x08.
2019-06-08 14:40:27 -07:00
Sam Lantinga
59483c6446 Fixed bug 4593 - Respect CMake's BUILD_SHARED_LIBS default behavior
tschwinger

Respect the BUILD_SHARED_LIBS variable when defined, and build either shared or static libs, which is CMake's default behavior (See https://cmake.org/cmake/help/latest/variable/BUILD_SHARED_LIBS.html).

If the variable is not defined, the current behavior remains unchanged and both variants are built where the platform supports it. This way, it remains possible to build both in one shot, which seems convenient for distro builds and useful to promote some consistency between them.
2019-06-08 14:36:03 -07:00
Sam Lantinga
68bb8d6c13 Fixed bug 4594 - Fix install location of CMake targets on Apple platforms
tschwinger

Followup to #3651

As already noted by Ryan, no framework is being built, so we better install to lib/cmake.

That code was originally part of a patch submitted by David Demelier, whose credit BTW got lost (I combined his patch for #3572 with fixes for #2576 and #3613 resulting in #3651 because things started to depend on another).

I tested that the configuration files are found correctly in the new location on MacOS X based on a hint to the root (see https://cmake.org/cmake/help/latest/command/find_package.html#search-procedure).
2019-06-08 14:34:38 -07:00
Sam Lantinga
d97387ccff Fixed bug 4600 - Dualshock 4 touchpad press is not detectable with SDL_JoystickGetButton
Dexter Friedman

When using a Dualshock 4 controller (model numbers CUH-ZCT1U and CUH-ZCT2U), pressing anywhere on the center touchpad does not send an SDL_JOYBUTTONDOWN event. I have verified this with testjoystick:

Repro steps:
1. Plug in a DS4 over USB
2. Compile testjoystick and run: testjoystick.exe 0
3. Press and hold the touchpad. Observe that no lime green box appears

Expected behavior:
A lime green box appears while the touchpad is pressed.

Notes:
I've attached a patch here that works on my PC and produces the expected behavior in testjoystick, for both DS4 model numbers I listed earlier.

If I understand correctly, by exposing this as a joystick button, the gamecontroller API mapping can be modified with a change to gamecontrollerdb.txt in the future.
2019-06-08 14:32:19 -07:00
Sam Lantinga
30a5e90ba8 Fixed bug 4603 - The iOS Test Xcode project file needs to add the metal framework
Dominik Reichardt

Similar to bug #4088 the iOS test suite needs the metal framework to link. Same linking error in the final linking phase.
2019-06-08 14:29:05 -07:00
Sam Lantinga
8a37848de9 Fixed bug 4605 - WASAPI_WaitDevice hang
Matt Brocklehurst

We've noticed that if you are playing audio on Windows via the WASAPI interface and you unplug and reconnect the device a few times the program hangs.

We've debugged the problem down to

static void
WASAPI_WaitDevice(_THIS)
{

   ... snip ...
 if (WaitForSingleObjectEx(this->hidden->event, INFINITE, FALSE) == WAIT_OBJECT_0) {
   ... snip ...
}

This WaitForSingleObjectEx does not havbe a time out defined, so it hangs there forever.

Our suggested fix we found was to include a time out of say 200mSec

We have done quite a bit of testing with this fix in place on various hardware configurations and it seems to have resolved the issue.
2019-06-08 13:41:46 -07:00
Zack Middleton
82af42761e hidapi: Use GameCube adapter controller port for player index
The Nintendo USB GameCube adapter has four controller ports. Return
the port number as 0 to 3 from SDL_JoystickGetPlayerIndex() and
SDL_JoystickGetDevicePlayerIndex().
2019-06-08 13:36:59 -07:00
Cameron Cawley
e7b514d8ff riscos: Fix iconv warnings 2019-01-13 23:36:31 +00:00
Sam Lantinga
15bae953b1 Fixed bug 4642 - Rework SDL_netbsdaudio to improve performance
Nia Alarie

The NetBSD audio driver has a few problems. Lots of obsolete code, and extremely bad performance and stuttering.

I have a patch in NetBSD's package system to improve it. This is my attempt to upstream it.

The changes include:

* Removing references to defines which are never used.
* Using the correct structures for playback and recording, previously they were the wrong way around.
* Using the correct types ('struct audio_prinfo' in contrast to 'audio_prinfo')
* Removing the use of non-blocking I/O, as suggested in #3177.
* Removing workarounds for driver bugs on systems that don't exist or use this driver any more.
* Removing all usage of SDL_Delay(1)
* Removing pointless use of AUDIO_INITINFO and tests that expect AUDIO_SETINFO to fail when it can't.

These changes bring its performance in line with the DSP audio driver.
2019-06-08 13:03:36 -07:00
Sam Lantinga
f2c8d8e9c4 Fixed bug 4443 - Incorrect scan code reported for numpad 5
bplu4t2f

When num lock is on, the scancode reported for numpad 5 is SDL_SCANCODE_KP_5, which is correct. However, when num lock is off, windows reports the VK_CLEAR virtual key code, which is incorrectly translated into SDL_SCANCODE_CLEAR inside of the VKeytoScancode(WPARAM vkey) function.
2019-06-08 10:47:43 -07:00
Cameron Gutman
5cb14f4467 Ignore Xbox One S gamepads with older firmware in HIDAPI
This older firmware had a different HID report style that we don't support.
2019-04-30 20:37:49 -07:00
Sam Lantinga
b1100ec18c Fixed Android build warning 2019-06-08 10:23:52 -07:00
Sam Lantinga
03cf24162f OpenSL ES audio cleanup and added a note with low latency audio discussion 2019-06-08 10:21:38 -07:00
Sam Lantinga
166d15fd75 Fixed surround sound channel setup for Android OpenSL ES audio driver 2019-06-07 15:09:15 -07:00
Sam Lantinga
23a2b47790 Protect against NULL device in the Android hidapi implementation 2019-06-07 09:00:26 -07:00
Sam Lantinga
9261e473d6 Potential fix for a crash we're seeing on Android that should in theory never happen. 2019-06-07 09:00:24 -07:00
Sam Lantinga
20ec866484 Added support for the Rotor Riot gamepad, and upcoming Xbox and PS4 controller support on iOS and tvOS
Patch contributed by Nat Brown
2019-06-06 08:20:53 -07:00
Sam Lantinga
6c0b304c93 Fixed bug 4655 - evdev is available on FreeBSD, check in 'configure' limited to Linux
Jan Martin Mikkelsen

The evdev interface is available on FreeBSD, with the xf86-input-evdev for include files in /usr/local/include/linux, so <linux/input.h> works, or when build with the native evdev option, where <dev/evdev/input.h> is available.
2019-06-05 12:03:45 -07:00
Sam Lantinga
23569362dc Use SDL sized types 2019-06-05 12:01:26 -07:00
Sam Lantinga
3e3294645f Fixed bug 4656 - SDL_evdev.c uses Linux specific integer types
Jan Martin Mikkelsen

The file src/core/linux/SDL_evdev.c uses the Linux specific types __u32 and __s32. This breaks things on FreeBSD when building with evdev.
2019-06-05 08:54:07 -07:00
Sam Lantinga
723d014336 Fixed bug 4171 - SDL_GetQueuedAudioSize is broken with WASAPI
Cameron Gutman

I was trying to use SDL_GetQueuedAudioSize() to ensure my audio latency didn't get too high while streaming data in from the network. If I get more than N frames of audio queued, I know that the network is giving me more data than I can play and I need to drop some to keep latency low.

This doesn't work well on WASAPI out of the box, due to the addition of GetPendingBytes() to the amount of queued data. As a terrible hack, I loop 100 times calling SDL_Delay(10) and SDL_GetQueuedAudioSize() before I ever call SDL_QueueAudio() to get a "baseline" amount that I then subtract from SDL_GetQueuedAudioSize() later. However, because this value isn't actually a constant, this hack can cause SDL_GetQueuedAudioSize() - baselineSize to be < 0. This means I have no accurate way of determining how much data is actually queued in SDL's audio buffer queue.

The SDL_GetQueuedAudioSize() documentation says: "This is the number of bytes that have been queued for playback with SDL_QueueAudio(), but have not yet been sent to the hardware." Yet, SDL_GetQueuedAudioSize() returns > 0 value when SDL_QueueAudio() has never been called.

Based on that documentation, I believe the current behavior contradicts the documented behavior of this function and should be changed in line with Boris's patch.

I understand that exposing the IAudioClient::GetCurrentPadding() value is useful, but a solution there needs to take into account what of that data is silence inserted by SDL and what is actual data queued by the user with SDL_QueueAudio(). Until that happens, I think the best approach is to remove the GetPendingBytes() call until SDL is able to keep track of queued data to make sense of it. This would make SDL_GetQueuedAudioSize() possible to use accurately with WASAPI.
2019-06-04 17:32:15 -07:00
Ryan C. Gordon
b5d3b6fc25 test: unify all the command line usage logging. 2019-05-28 17:39:13 -04:00
Alex Szpakowski
9b5811592d iOS: return SDL_GetWindowSize from SDL_GL_GetDrawableSize if there's no GLES view in the window (matches the behaviour of SDL_GL_GetDrawableSize on other platforms). Addresses bug #4629. 2019-05-26 18:53:36 -03:00
Sam Lantinga
dce56ab9fa Added a function to get the current Android SDK version at runtime 2019-05-23 14:19:00 -07:00
Sam Lantinga
f3e76ea1d0 Use the OpenSL ES audio driver by default on Android, as it has the lowest latency. 2019-05-23 13:47:30 -07:00
Sam Lantinga
02f9667a08 Fixed static and buzzing when trying to use floating point audio on the OpenSL ES audio driver. 2019-05-23 13:47:27 -07:00
Sam Lantinga
e401b95025 Return an error if both mouse relative mode and mouse warping are unavailable, instead of asserting. 2019-05-23 11:32:36 -07:00
Sam Lantinga
8b57331e71 Fixed hiding the Android virtual keyboard when the return key is pressed 2019-05-23 11:05:43 -07:00
Sylvain Becker
f91b87859c Android: minimum size for IME, so that it takes focus
In API 28, 0 width views can't take focus, so if someone tries to position the IME without setting a width, they'll stop getting text events.

Tested on Android 9: with a 0 size, it would send correctly letters a, b, c, etc. but not numbers.
2019-05-23 09:08:40 +02:00
Sam Lantinga
7ec514d48f Improved iOS Bluetooth keyboard support
* Don't stop text input after the return key is pressed
* Handle arrow and escape keys
2019-05-22 17:39:51 -07:00
Sam Lantinga
50aab19877 Fixed bug 4639 - CMake build does not generate libhidapi.so for Android
Manuel Sabogal

I noticed that the current Android.mk builds a libhidapi.so library for Android but the CMake build hasn't been updated to do so. I'll attach a patch that fixes this issue.
2019-05-21 17:33:31 -07:00
Cameron Gutman
9b2202828a Fix use-after-free when pumping the event loop after SDL_DestroyWindow()
Closing the window is asynchronous, but we free the window data immediately,
so we can get an updateLayer callback before the window is really destroyed which
will cause us to access the freed memory.

Clearing the content view will cause it to be immediately released, so no further
updateLayer callbacks will occur.
2019-04-28 17:37:49 -07:00
Sam Lantinga
a4e33b9cc4 Added support for Bluetooth keyboards on iOS
In this case the keyboard is shown and immediately hidden, but we still want to accept text input
2019-05-20 14:31:03 -07:00
Sam Lantinga
582a3c994d Fixed mouse focus for touch events on iOS 2019-05-20 14:08:35 -07:00
Ryan C. Gordon
4bd808346a vulkan: Swapped out a free() that should have been an SDL_free().
Fixes (for real this time!) the Visual Studio builds.
2019-05-20 00:41:18 -04:00
Ryan C. Gordon
ebbb295e85 vulkan: Patched to compile on Visual Studio. 2019-05-19 23:29:50 -04:00
Ryan C. Gordon
d778b26dd1 Patched to compile in C89 mode. 2019-05-19 20:25:02 -04:00