|
9f56c7cf
|
2022-07-20T16:39:19
|
|
audio: Remove 5.1->X SIMD converters, add SSE mono->stereo.
The 5.1 versions didn't use the new algorithm, and making that new
algorithm work took so many permutes that it was significantly slower
than just using the scalar versions.
However, mono-to-stereo is an extremely common conversion, and it's
trivial to accelerate it with plain SSE, so that was added!
|
|
|
b83ae9f2
|
2022-07-19T22:40:51
|
|
audio: Replaced some debug-printfs with debug-SDL_Logs. :)
|
|
|
49ec8db5
|
2022-07-19T22:04:49
|
|
audio: Generate the channel converter code from a program.
|
|
|
f06cc3e9
|
2022-07-19T22:03:56
|
|
audio: "SL" means "surround left" not "side left", etc.
|
|
|
5a0c8198
|
2022-07-19T22:03:02
|
|
audio: Add channel convert filter _after_ choosing an SIMD version.
|
|
|
25727790
|
2022-07-19T02:16:08
|
|
audio: first attempt at rewriting the channel converters.
This is not ready for production use!
|
|
|
29694869
|
2022-04-28T15:56:52
|
|
audio: Revert one of the resampler optimizations.
This is the one that splits the "left wing" into two for loops to
bubble out the conditional that decides if it should read from the
left padding or the input buffer.
I still believe the optimization is good, but the basic logic of it
was incorrect, and needs to be reexamined and fixed before going
back into revision control.
|
|
|
111c3add
|
2022-04-10T13:23:51
|
|
audio: Resampler optimizations.
- Calculate `j * RESAMPLER_SAMPLES_PER_ZERO_CROSSING` once per loop
iteration since we use it multiple times.
- Do the left-wing loop in two sections: while `srcframe < 0` and then
the remaining calculations when `srcframe >= 0`. This bubbles a conditional
out of every iteration of a tight loop, giving us a boost. We could
_probably_ do this to the right-wing loop too, but it's less straightforward
there.
- The real win: Use floats instead of doubles. This almost doubles the speed
of the entire function on Intel CPUs, and for embedded things without
hardware-level support for doubles, the speedup is enormous. This in
theory might reduce audio quality, though, and I had to put a check in
place to avoid a division-by-zero that we avoided at higher precision, but
this is likely to be worth keeping for at least the Sony PSP and other
smaller platforms, if not everyone.
|
|
|
de019568
|
2022-04-09T23:43:57
|
|
audio: Prebake the resampler's kaiser table instead of doing it at runtime.
|
|
|
727eef70
|
2022-04-09T10:12:49
|
|
audio: SDL_ConvertStereoToMono_SSE3 missed an unaligned load.
|
|
2c6a9c51
|
2022-03-16T09:36:43
|
|
minor optimization (SDL_audiocvt.c)
|
|
5905696e
|
2022-03-15T23:10:04
|
|
SDL_audiocvt.c: minor cleanup.
|
|
7c421fec
|
2022-03-14T05:55:41
|
|
SDL_audiocvt.c: Don't byteswap 8-bit streams
Otherwise, this results an assert on big endian machines when attenpting to use SDL_LoadWAV_RW function to load 8-bit WAV files.
|
|
|
f91211eb
|
2022-01-19T14:51:42
|
|
cleanup WASAPI_PrepDevice
- reorganize the loop which checks for the right wave-format
- use the return value of UpdateAudioStream
- ensure SetError is called in SDL_NewAudioStream
|
|
|
ebdd5366
|
2022-01-17T16:26:02
|
|
use SDL_InvalidParamError or SDL_assert instead of custom SDL_SetError
|
|
120c76c8
|
2022-01-03T09:40:00
|
|
Updated copyright for 2022
|
|
5be8a221
|
2021-11-17T20:37:51
|
|
Add audio conversion from/to 61
|
|
|
b649314d
|
2021-11-15T13:43:40
|
|
Add comment to audio clipping (see bug #4104)
|
|
|
072e3fdf
|
2021-10-14T23:17:08
|
|
Fixed bug #4534: NEON implementation of Convert51ToStereo (Thanks Ryan!)
|
|
1e921352
|
2021-08-13T07:58:48
|
|
Improve portability of SDL_Convert51ToStereo_AVX
Don't rely on checking __clang_major__ since it is not comparable
between different vendors. Don't use "#pragma clang attribute" since it
is only available in relatively recent versions, there's no obvious way
to check if it's supported, and just using __attribute__ directly (for
gcc as well) results in simpler code anyway.
|
|
|
44a76710
|
2021-08-11T12:55:50
|
|
Don't disable clang avx instrinsics on win32 if __AVX__ is defined.
C.f.: https://github.com/libsdl-org/SDL/issues/4533
|
|
|
bec78357
|
2021-08-06T14:20:55
|
|
Better fix for compiling using clang on Windows
This fixes https://github.com/libsdl-org/SDL/issues/4533
|
|
|
cae7bd9b
|
2021-08-06T12:01:24
|
|
Don't use AVX with clang if the compiler isn't building with AVX instructions
Fixes https://github.com/libsdl-org/SDL/issues/4533
|
|
|
659e1f0a
|
2021-07-29T17:49:52
|
|
audiocvt: The to-5.1 converters now soften FL and FR channels more.
This is experimental and might be tweaked further.
Reference #4104.
Also reference:
https://github.com/Keriew/augustus/issues/194#issuecomment-847655049
|
|
8f38ba4d
|
2021-07-29T18:02:47
|
|
Fix casts that should be using uintptr_t
This is needed to support CHERI, and thus Arm's experimental Morello
prototype, where pointers are implemented using unforgeable capabilities
that include bounds and permissions metadata to provide fine-grained
spatial and referential memory safety, as well as revocation by sweeping
memory to provide heap temporal memory safety.
On most systems (anything with a flat memory hierarchy rather than using
segment-based addressing), size_t and uintptr_t are the same type.
However, on CHERI, size_t is just an integer offset, whereas uintptr_t
is still a capability as described above. Casting a pointer to size_t
will strip the metadata and validity tag, and casting from size_t to a
pointer will result in a null-derived capability whose validity tag is
not set, and thus cannot be dereferenced without faulting.
The audio and cursor casts were harmless as they intend to stuff an
integer into a pointer, but using uintptr_t is the idiomatic way to do
that and silences our compiler warnings (which our build tool makes
fatal by default as they often indicate real problems). The iconv and
egl casts were true positives as SDL_iconv_t and iconv_t are pointer
types, as is NativeDisplayType on most OSes, so this would have trapped
at run time when using the round-tripped pointers. The gles2 casts were
also harmless; the OpenGL API defines this argument to be a pointer type
(and uses the argument name "pointer"), but it in fact represents an
integer offset, so like audio and cursor the additional idiomatic cast
is needed to silence the warning.
|
|
53987e9b
|
2021-07-28T21:03:42
|
|
Optimized SDL_Convert51ToStereo_AVX
|
|
|
cf85710c
|
2021-07-28T22:55:10
|
|
SDL_audiocvt.c: disable AVX for clang < 5 and gcc < 4.9
See: https://github.com/libsdl-org/SDL/issues/4533
|
|
|
8d790b10
|
2021-07-27T12:23:46
|
|
audiocvt: stereo-to-mono SSE3 now uses unaligned accesses.
On modern CPUs, there's no penalty for using the unaligned instruction on
aligned memory, but now it can vectorize unaligned data too, which even if
it's not optimal, is still going to be faster than the scalar fallback.
Fixes #4532.
|
|
b2c8d3e9
|
2021-06-11T22:14:54
|
|
audiocvt: 5.1 to Stereo conversion utilizing AVX
|
|
|
638befc1
|
2021-06-10T17:22:39
|
|
audiocvt: 5.1 to Stereo conversion utilizing SSE
|
|
|
db56526f
|
2021-06-10T13:07:27
|
|
audiocvt: Replace divisions
Division is expensive and the compiler can not replace it themselves.
Though we now we will be ok in terms of floating point edge behaviour.
|
|
|
58884e4c
|
2021-05-04T00:23:40
|
|
SDL_audiocvt.c: fixed MSVC double->float conversion warnings.
|
|
|
f8695185
|
2021-04-27T11:07:51
|
|
Audio: normalize conversion Stereo to 5.1, Quad to 7.1, 5.1 to 7.1 (bug #4104)
|
|
|
21349901
|
2021-04-27T10:57:48
|
|
Audio: convert 5.1 to 7.1, use right-surround for r-front and r-back (see #4104)
|
|
|
9130f7c3
|
2021-01-02T10:25:38
|
|
Updated copyright for 2021
|
|
|
cb361896
|
2020-12-09T07:16:22
|
|
Fixed bug 5235 - All internal sources should include SDL_assert.h
Ryan C. Gordon
We should really stick this in SDL_internal.h or something so it's always available.
|
|
|
a8780c6a
|
2020-01-16T20:49:25
|
|
Updated copyright date for 2020
|
|
|
3ce56f62
|
2020-01-13T08:12:10
|
|
Fixed error formatting
|
|
|
e3cedf96
|
2020-01-11T04:38:13
|
|
Add the destination format to the error when conversion isn't possible
|
|
|
7a47c292
|
2019-07-31T01:22:02
|
|
Fix bug 4746 - introduce SDL_zeroa macro.
|
|
|
a21b5b30
|
2019-06-08T19:09:43
|
|
Fixed build
|
|
|
31765242
|
2019-06-08T18:22:18
|
|
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.
|
|
|
5e13087b
|
2019-01-04T22:01:14
|
|
Updated copyright for 2019
|
|
|
4773690d
|
2018-06-25T12:55:23
|
|
Deal with possible malloc(0) calls, as pointed out by static analysis.
|
|
|
e3cc5b2c
|
2018-01-03T10:03:25
|
|
Updated copyright for 2018
|
|
|
e830ef34
|
2017-10-20T16:53:42
|
|
Fixed typo converting 4 channel audio to 2 channel
|
|
|
9a291c1e
|
2017-10-20T14:51:22
|
|
Added a note about adjusting channel weights when converting to fewer channels
|
|
|
72932906
|
2017-10-19T18:05:42
|
|
audio: Added SDL_AudioStreamFlush().
|
|
|
e98920f5
|
2017-10-18T23:49:46
|
|
Check correct variable for malloc() results.
|
|
|
653ab5d9
|
2017-10-18T19:26:36
|
|
Added a staging buffer to the audio stream so that we can accumulate small amounts of data if needed when resampling
|
|
|
80f8464d
|
2017-10-18T15:54:05
|
|
Added audio stream conversion functions:
SDL_NewAudioStream
SDL_AudioStreamPut
SDL_AudioStreamGet
SDL_AudioStreamAvailable
SDL_AudioStreamClear
SDL_FreeAudioStream
|
|
|
ba10d2b6
|
2017-10-12T13:55:35
|
|
Fixed compiler warning
|
|
|
5e5f2290
|
2017-10-11T12:07:43
|
|
audio: Turns out the accumulation errors sound better. :/
Moving to double fixed the overflows, but using "time = i * incr" instead of
"time += incr" causes clicks in the output.
|
|
|
9bd2c6b4
|
2017-10-11T11:51:14
|
|
audio: Moved the resampler state up to double precision.
Fixes more buffer overflows.
|
|
|
b2f5123b
|
2017-10-11T11:43:35
|
|
audio: calculate resampling time directly, don't increment (thanks, Eric!).
Fixes buffer overruns as floating point errors accumulate.
Partially fixes Bugzilla #3848.
|
|
|
763c3871
|
2017-10-11T02:33:55
|
|
audio: clamp resampler interpolation values to prevent buffer overflow.
Partially fixes Bugzilla #3848.
|
|
|
0085f917
|
2017-10-11T02:31:58
|
|
audio: Moved unchanging variable out of loop.
|
|
|
cb8bf6bb
|
2017-10-11T02:03:05
|
|
audio: Make sure audio stream resampling doesn't overflow buffers.
|
|
|
459e2b0b
|
2017-10-11T01:37:11
|
|
audio: Fixed check for minimum audio stream put size.
|
|
|
903ff641
|
2017-10-10T22:31:02
|
|
audio: SDL_ResampleCVT() should use memmove instead of memcpy.
This copy can overlap.
Fixes Bugzilla #3849.
|
|
|
42fff7ce
|
2017-10-10T22:18:46
|
|
audio: Don't stack-allocate resampler padding.
(I thought padding size ranged from 5 frames to ~30 frames (based around
RESAMPLER_ZERO_CROSSINGS, which is 5), but it's actually between 512 and
several thousands (based on RESAMPLER_SAMPLES_PER_ZERO_CROSSING)). It gets
big fast when downsampling.
|
|
|
37d89aa1
|
2017-10-10T16:12:56
|
|
audio: reworked audio streams to have right-hand resampling padding available.
Fixes Bugzilla #3851.
|
|
|
099ae43e
|
2017-09-22T22:28:21
|
|
audio: Fixed compiler warning on Visual Studio.
|
|
|
d74c00e6
|
2017-09-22T08:51:45
|
|
Fixed memory leak when HAVE_ALLOCA isn't defined
|
|
|
6d206a7b
|
2017-09-22T07:42:24
|
|
audio: Stream resampling now saves some samples from previous run for padding.
Previously, the padding was silence, which was a problem when streaming since
you would sample a little bit of this silence between each buffer.
We still need a means to get padding data for the right hand side, but this
patch makes the resampler output more correct.
|
|
|
1a3b95a1
|
2017-09-21T02:51:14
|
|
audio: Replaced the resampler. Again.
This time it's using real math from a real whitepaper instead of my previous
amateur, fast-but-low-quality attempt. The new resampler does "bandlimited
interpolation," as described here: https://ccrma.stanford.edu/~jos/resample/
The output appears to sound cleaner, especially at high frequencies, and of
course works with non-power-of-two rate conversions.
There are some obvious optimizations to be done to this still, and there is
other fallout: this doesn't resample a buffer in-place, the 2-channels-Sint16
fast path is gone because this resampler does a _lot_ of floating point math.
There is a nasty hack to make it work with SDL_AudioCVT.
It's possible these issues are solvable, but they aren't solved as of yet.
Still, I hope this effort is slouching in the right direction.
|
|
|
d619d885
|
2017-08-28T21:42:39
|
|
Fixed bug 3662 - Error message when using the audio conversion setup without an initialized audio subsystem is a bit vague
Simon Hug
This issue actually raises the question if this API change (requirement of initialized audio subsystem) is breaking backwards compatibility. I don't see the documentation saying it is needed in 2.0.5.
|
|
|
b128e880
|
2017-08-29T00:41:45
|
|
audio: A whole bunch of improvements to audio conversion (thanks, Solra!).
"Major changes, roughly in order of appearance:
- Use float math everywhere, instead of promoting to double and casting back
all the time.
- Conserve sound energy when downmixing any channel into two other channels.
- Add a QuadToStereo filter. (The previous technique of reusing StereoToMono
never worked, since it assumed an incorrect channel layout for 4.0.)
- Add a 71to51 filter. This removes just under half of the cases the previous
code would silently break in.
- Add a QuadTo51 filter. More silent breakage fixed.
- Add a 51to71 filter, removing another almost-half of the silently broken
cases.
- Add 8 to the list of values SDL_SupportedChannelCount will accept.
- Change SDL_BuildAudioCVT's channel-related logic to handle every case, and
to actually fail if it fails instead of silently corrupting sound data and/or
crashing down the road."
(Note that SDL doesn't otherwise support 7.1 audio yet, but hopefully it will
soon and the 7.1 converters are an important piece of that. --ryan.)
Fixes Bugzilla #3727.
|
|
|
e3e6b4fd
|
2017-08-18T16:52:19
|
|
audio: better docs on conversion APIs, error if not init'd (thanks, Simon!).
Fixes Bugzilla #3662.
|
|
|
1683a0c1
|
2017-07-05T12:04:37
|
|
audio: trying to pacify static analysis.
|
|
|
a509719f
|
2017-06-12T21:35:24
|
|
audio: Converter now checks a strict list of channels and formats we support.
|
|
|
553b3286
|
2017-06-12T16:39:15
|
|
Fixed bug 3668 - Overflow of SDL_AudioCVT.filters with some downmixes
Simon Hug
There's a chance that an audio conversion from many channels to a few can use more than 9 audio filters. SDL_AudioCVT has 10 SDL_AudioFilter pointers of which one has to be the terminating NULL pointer. The SDL code has no checks for this limit. If it overflows there can be stack or heap corruption or a call to 0xa.
Attached patch adds a function that checks for this limit and throws an error if it is reached. Also adds some documentation.
Test parameters that trigger this issue:
AUDIO_U16MSB with 224 channels at 46359 Hz
V
AUDIO_S16MSB with 6 channels at 27463 Hz
The fuzzer program I uploaded in bug 3667 has more of them.
|
|
|
ca0bf151
|
2017-03-03T16:38:17
|
|
Fix some more compiler warnings on armcc.
|
|
|
d526b8a1
|
2017-03-02T13:33:04
|
|
Some patches to make SDL compile with armcc (ARM's C compiler).
|
|
|
ad9c702f
|
2017-02-13T16:56:41
|
|
audio: SDL_AudioStream's *_sample_frame_size should be in bytes, not bits.
Fixes failures where SDL_AudioStreamGet() incorrectly thinks it got a partial
sample frame request.
|
|
|
47e2f4e9
|
2017-01-24T20:30:48
|
|
audio: libsamplerate can't resample in-place; make space for a copy if needed.
|
|
|
c7f9dcb6
|
2017-01-24T15:52:22
|
|
audio: Offer a hint for libsamplerate quality/speed tradeoff.
This defaults to the internal SDL resampler, since that's the likely default
without a system-wide install of libsamplerate, but those that need more can
tweak this.
|
|
|
1da3a337
|
2017-01-24T10:09:29
|
|
audio: Fix static analysis concerns about a dead assignment.
|
|
|
8f627c1c
|
2017-01-24T00:51:33
|
|
audio: Make sure SDL_AudioStream's work buffer is 16-byte aligned, for SIMD.
Note the giantic FIXME, though!
|
|
|
17dcee20
|
2017-01-24T00:17:40
|
|
audio: Streams now resample in-place. Removed second allocated buffer.
|
|
|
b5eeab77
|
2017-01-24T00:08:24
|
|
audio: allow stereo Sint16 resampling fast path in SDL_AudioStream.
This currently favors libsamplerate over the fast path (quality over speed),
but I'm not sure that's the correct approach, as there may be surprising
changes in performance metrics depending on what packages are available on
a user's system. That being said, currently, the only thing with access to
SDL_AudioStream is an SDL audio device's thread, and it might be mostly idle
otherwise, so maybe this is generally good.
|
|
|
a80cb672
|
2017-01-24T00:03:36
|
|
audio: Fixed off-by-one error in upsampling.
|
|
|
dad07f96
|
2017-01-23T16:45:50
|
|
audio: Resampler now special-cases stereo and mono processing.
Turns out that iterating from 0 to channels-1 was a serious performance hit!
These cases now tend to match or beat the original audio resampler's speed!
|
|
|
8ce6ddf1
|
2017-01-23T16:42:47
|
|
audio: Fixed incorrect pointer in SDL_ResampleCVT_si16_c2().
Forgot to update this when we changed this to process in-place. Whoops!
|
|
|
3594bf8e
|
2017-01-23T01:05:44
|
|
audio: Wired up new SSE code to build system.
|
|
|
64056e81
|
2017-01-23T00:57:19
|
|
audio: Added SSE3 implementation of SDL_ConvertStereoToMono().
|
|
|
8855daac
|
2017-01-22T23:48:15
|
|
audio: Make the simple resampler operate in-place.
This allows us to avoid an extra copy, allocate less memory and reduce cache
pressure. On the downside: we have to do a lot of tapdancing to resample the
buffer in reverse when the output is growing.
|
|
|
202ab30c
|
2017-01-22T20:27:48
|
|
audio: Special case for resampling stereo AUDIO_S16SYS audio data.
This is a fairly common case, so we avoid the conversion to/from float here.
|
|
|
a7f86f2f
|
2017-01-22T20:18:59
|
|
audio: don't cast to double in SDL_ConvertStereoToMono().
It's expensive and (hopefully) unnecessary. If this becomes an overflow
problem, we could multiply both values by 0.5f before adding them, but let's
see if we can get by without the extra multiplication first.
|
|
|
83454c82
|
2017-01-20T16:26:24
|
|
audio: removed conditional from simple resampler's inner loop.
We never seem to overflow the source buffer now; this might have been a
leftover from a bug that was covered by Vitaly's fixes?
Removing this conditional makes the resampler 10-20% faster. Left an
assert in there for debug builds, in case this still happens.
|
|
|
3e1679c8
|
2017-01-18T02:11:56
|
|
audio: Several fixes to "simple" resampler (thanks, Vitaly!).
Fixes Bugzilla #3551.
|
|
|
1e66d457
|
2017-01-15T05:01:59
|
|
audio: Some fixes to the audio data type converter code.
Removed some needless things ("len / sizeof (Uint8)"), and made sure the
int32 -> float code uses doubles to avoid working with large integer values
in a 32-bit float.
|
|
|
bf11cd50
|
2017-01-09T20:37:52
|
|
Fixed bug 3552 - Building SDL in release mode fails under VS 2017 RC
Lukasz Biel
Tried to compile SDL2 using newest version of VS.
Got:
SDL_audiocvt.obj : error LNK2019: unresolved external symbol memcpy referenced in function SDL_ResampleCVT
1>E:\Users\dotPo\Lib\SDL\VisualC\x64\Release\SDL2.dll : fatal error LNK1120: 1 unresolved externals
whole compilation process: http://pastebin.com/eWDAvBce
Steps to reproduce:
clone http://hg.libsdl.org/SDL using tortoise hg,
open SDL\VisualC\SDL.sln,
when promted if should retarget solution click ok,
select release x64 build type,
Build/Build Solution
attempt 2, using Visual Studio cmake support:
open folder SDL\
select release x64 build type,
run CMake\Build CMakeLists.txt
build fails
When switched to debug build type, buils succeeds in both cases.
VS 2017 is still beta.
|
|
|
23020f92
|
2017-01-09T16:31:57
|
|
audio: Don't ever use libsamplerate in the SDL_AudioCVT codepath.
It causes audio pops if you're converting in chunks (and needs to
allocate/initialize/free on each convert). We'll either adjust this interface
when we break ABI for 2.1 to make this usable, or publish the SDL_AudioStream
API for those that want a streaming solution.
In the meantime, the "simple" resampler produces "good enough" audio without
pops and doesn't have to be initialized, so that'll do for now on the
SDL_AudioCVT interface.
|
|
|
063c9d40
|
2017-01-09T06:00:58
|
|
audio: Replaced older resamplers in SDL_AudioCVT with the new ones.
|
|
|
38854e03
|
2017-01-08T16:18:49
|
|
audio: Improvements in channel conversion code.
|
|
|
d005dc21
|
2017-01-08T14:23:15
|
|
audio: Patched to compile with libsamplerate support.
|
|
|
19e937fc
|
2017-01-08T14:18:03
|
|
audio: libsamplerate loading now happens once at init time.
|
|
|
98cc9d10
|
2017-01-08T14:17:09
|
|
Fixed coding style on a function signature.
|
|
|
df25258a
|
2017-01-06T20:43:53
|
|
Added configure and cmake support for libsamplerate
|