[Libav-user] I can't get audio decoding to work.
Bill Messenger
apothemmusic at gmail.com
Sun Jun 26 03:39:23 CEST 2016
But I tried "std::memcpy(sampleBuffer + totalBufferSize, frame->data[0],
dataSize-4);", and it still gave me the same result. I even tried
subtracting 8, 16, 32, 64, and so on, and it still gave the same result.
When I go up to dataSize/2, however, it doesn't give me that UNADDRESSABLE
ACCESS error. Does frame->data[0] only contain half of the data for planar
audio?
On Fri, Jun 24, 2016 at 6:58 PM, Bill Messenger <apothemmusic at gmail.com>
wrote:
> I used "_heapchk" in every place I could think of in my function, and it
> always returns "_HEAPOK" before the line that crashes. I can't see the
> value of _heapchk() after the crash though, Visual Studio won't let me
> continue after the crash. I also made a new project that just uses the
> FFmpeg libraries to make sure the problem wasn't with PortAudio or
> wxWidgets, but it's still crashing in that same place. When I ran it
> through Dr. Memory with an mp3 or ogg file, it output an UNADDRESSABLE
> ACCESS error. When I ran it with a wav or flac file, it didn't output that
> error.
>
> This is what Dr. Memory output with the mp3 file one: (Line 154 of
> audiodecoder.cpp is the "std::memcpy(sampleBuffer + totalBufferSize,
> frame->data[0], dataSize);" that was crashing before. I guess I should also
> mention that the crash happens in both 32 bit and 64 bit builds in both
> debug and release mode.)
>
> Dr. Memory version 1.10.1 build 3 built on Apr 10 2016 18:05:55
> Dr. Memory results for pid 2804: "FFmpegTest.exe"
> Application cmdline:
> "C:\Users\BillM\Documents\My_Files\Code\FFmpegTest\x64\Release\FFmpegTest.exe"
> Recorded 115 suppression(s) from default C:\Program Files (x86)\Dr.
> Memory\bin64\suppress-default.txt
>
> Error #1: UNADDRESSABLE ACCESS beyond heap bounds: reading
> 0x000002a77c94fbb8-0x000002a77c94fbbc 4 byte(s)
> # 0 replace_memmove
> [d:\drmemory_package\drmemory\replace.c:763]
> # 1 AudioDecoder::decodeFile
> [c:\users\billm\documents\my_files\code\ffmpegtest\ffmpegtest\audiodecoder.cpp:154]
> # 2 VCRUNTIME140.dll!set_se_translator +0x47a (0x00007ff846f44acb
> <VCRUNTIME140.dll+0x4acb>)
> # 3 VCRUNTIME140.dll!set_se_translator +0x21b (0x00007ff846f4486c
> <VCRUNTIME140.dll+0x486c>)
> # 4 MSVCP140.dll!std::basic_ostream<>::flush
> [f:\dd\vctools\crt\crtw32\stdhpp\ostream:581]
> # 5 KERNEL32.dll!BaseThreadInitThunk +0x21 (0x00007ff850678102
> <KERNEL32.dll+0x18102>)
> Note: @0:00:01.844 in thread 11544
> Note: refers to 1 byte(s) beyond last valid byte in prior malloc
> Note: prev lower malloc: 0x000002a77c94f290-0x000002a77c94fbb7
> Note: allocated here:
> Note: # 0 replace_malloc
> [d:\drmemory_package\common\alloc_replace.c:2576]
> Note: # 1 avcodec-57.dll!? +0x0 (0x00007ff81b56e61c
> <avcodec-57.dll+0x4fe61c>)
> Note: # 2 avutil-55.dll!? +0x0 (0x00007ff821617b6c
> <avutil-55.dll+0x27b6c>)
> Note: # 3 avutil-55.dll!? +0x0 (0x00007ff8215f627a
> <avutil-55.dll+0x627a>)
> Note: # 4 avutil-55.dll!? +0x0 (0x00007ff821617e92
> <avutil-55.dll+0x27e92>)
> Note: # 5 avutil-55.dll!? +0x0 (0x00007ff8215f6c06
> <avutil-55.dll+0x6c06>)
> Note: # 6 avutil-55.dll!? +0x0 (0x00007ff8215f69a8
> <avutil-55.dll+0x69a8>)
> Note: # 7 avcodec-57.dll!? +0x0 (0x00007ff81b7af025
> <avcodec-57.dll+0x73f025>)
> Note: # 8 avcodec-57.dll!? +0x0 (0x00007ff81c00dc4d
> <avcodec-57.dll+0xf9dc4d>)
> Note: # 9 avcodec-57.dll!? +0x0 (0x00007ff81b7af385
> <avcodec-57.dll+0x73f385>)
> Note: #10 avcodec-57.dll!? +0x0 (0x00007ff81b7af66b
> <avcodec-57.dll+0x73f66b>)
> Note: #11 avcodec-57.dll!? +0x0 (0x00007ff81b574b9a
> <avcodec-57.dll+0x504b9a>)
> Note: instruction: mov (%rcx) -> %ecx
>
> Error #2: LEAK 1175 direct bytes 0x000002a77c946100-0x000002a77c946597 + 0
> indirect bytes
> # 0 replace_malloc
> [d:\drmemory_package\common\alloc_replace.c:2576]
> # 1 avutil-55.dll!? +0x0
> (0x00007ff82161c761 <avutil-55.dll+0x2c761>)
> # 2 avutil-55.dll!? +0x0
> (0x00007ff821617b6c <avutil-55.dll+0x27b6c>)
> # 3 avcodec-57.dll!? +0x0
> (0x00007ff81b646873 <avcodec-57.dll+0x5d6873>)
> # 4 AudioDecoder::decodeFile
> [c:\users\billm\documents\my_files\code\ffmpegtest\ffmpegtest\audiodecoder.cpp:63]
> # 5 VCRUNTIME140.dll!set_se_translator +0x47a
> (0x00007ff846f44acb <VCRUNTIME140.dll+0x4acb>)
> # 6 VCRUNTIME140.dll!set_se_translator +0x21b
> (0x00007ff846f4486c <VCRUNTIME140.dll+0x486c>)
> # 7 MSVCP140.dll!std::basic_filebuf<>::_Unlock
> [f:\dd\vctools\crt\crtw32\stdhpp\fstream:365]
> # 8 MSVCP140.dll!std::basic_ostream<>::flush
> [f:\dd\vctools\crt\crtw32\stdhpp\ostream:581]
> # 9 KERNEL32.dll!BaseThreadInitThunk +0x21
> (0x00007ff850678102 <KERNEL32.dll+0x18102>)
>
> ===========================================================================
> FINAL SUMMARY:
>
> DUPLICATE ERROR COUNTS:
> Error # 1: 1706
>
> SUPPRESSIONS USED:
>
> ERRORS FOUND:
> 1 unique, 1706 total unaddressable access(es)
> 0 unique, 0 total invalid heap argument(s)
> 0 unique, 0 total GDI usage error(s)
> 0 unique, 0 total handle leak(s)
> 0 unique, 0 total warning(s)
> 1 unique, 1 total, 1175 byte(s) of leak(s)
> 0 unique, 0 total, 0 byte(s) of possible leak(s)
> ERRORS IGNORED:
> 1 potential leak(s) (suspected false positives)
> (details: C:\Users\BillM\AppData\Roaming\Dr.
> Memory\DrMemory-FFmpegTest.exe.2804.000\potential_errors.txt)
> 14 unique, 14 total, 7286 byte(s) of still-reachable
> allocation(s)
> (re-run with "-show_reachable" for details)
> Details: C:\Users\BillM\AppData\Roaming\Dr.
> Memory\DrMemory-FFmpegTest.exe.2804.000\results.txt
>
>
> On Wed, Jun 22, 2016 at 3:58 PM, Bill Messenger <apothemmusic at gmail.com>
> wrote:
>
>> I added some code to check if it was crashing because dataSize is
>> negative. It's not. Also, I found out that dtSize in this instance is 4608,
>> and no matter what I do, sizeof(frame->data[0]) is always 8. Also, I tried
>> sizeof(frame->data) which returned 64.
>>
>> int dtSize = av_samples_get_buffer_size(nullptr, codecCtx->channels,
>> frame->nb_samples, codecCtx->sample_fmt, 1);
>> wxGetApp().popUpErrorDialog("dtSize: " + std::to_string(dtSize));
>> dataSize = dtSize;
>>
>> while(totalBufferSize + dataSize > estimatedBuffSize)
>> {
>> estimatedBuffSize *= 1.1;
>> sampleBuffer = (uint8_t*)std::realloc(sampleBuffer, estimatedBuffSize);
>> }
>> wxGetApp().popUpErrorDialog("sizeof(frame->data[0]): " +
>> std::to_string(sizeof(frame->data[0])));
>>
>> std::memcpy(sampleBuffer + totalBufferSize, frame->data[0], dataSize);
>>
>> totalBufferSize += dataSize;
>> totalSamples += frame->nb_samples;
>>
>> On Mon, Jun 20, 2016 at 5:47 PM, Bill Messenger <apothemmusic at gmail.com>
>> wrote:
>>
>>> This is the function I'm using to decode the audio file. My guess is
>>> that it's crashing because "dataSize" is larger than "frame->data[0]", but
>>> I'm pretty sure my calculation of dataSize is correct. Am I copying the
>>> frame data to "sampleBuffer" the wrong way?
>>>
>>> bool AudioDecoder::decodeFile(std::string* filename)
>>> {
>>> reset(); // if a file has already been decoded, free it from memory and
>>> reset
>>>
>>> AVFormatContext* formatCtx = avformat_alloc_context();
>>> if(avformat_open_input(&formatCtx, filename->c_str(), nullptr, nullptr)
>>> < 0)
>>> {
>>> wxGetApp().popUpErrorDialog("Couldn't open \"" + *filename + "\".");
>>> avformat_close_input(&formatCtx);
>>> return false;
>>> }
>>>
>>> if(avformat_find_stream_info(formatCtx, nullptr) < 0)
>>> {
>>> wxGetApp().popUpErrorDialog("Couldn't find file info for \"" + *filename
>>> + "\".");
>>> avformat_close_input(&formatCtx);
>>> return false;
>>> }
>>>
>>> av_dump_format(formatCtx, 0, filename->c_str(), false);
>>>
>>> int streamID = -1;
>>> for(int i = 0; i < formatCtx->nb_streams; i++)
>>> {
>>> if(formatCtx->streams[i]->codec->codec_type == AVMEDIA_TYPE_AUDIO)
>>> {
>>> streamID = i;
>>> break;
>>> }
>>> }
>>> if(streamID == -1)
>>> {
>>> wxGetApp().popUpErrorDialog("\"" + *filename + "\" does not contain
>>> audio.");
>>> avformat_close_input(&formatCtx);
>>> return false;
>>> }
>>>
>>> AVCodecContext* codecCtx = formatCtx->streams[streamID]->codec;
>>> AVCodec* codec = avcodec_find_decoder(codecCtx->codec_id);
>>> if(!codec)
>>> {
>>> wxGetApp().popUpErrorDialog("Couldn't find the codec for \"" + *filename
>>> + "\".");
>>> avformat_close_input(&formatCtx);
>>> return false;
>>> }
>>> if(avcodec_open2(codecCtx, codec, nullptr) < 0)
>>> {
>>> wxGetApp().popUpErrorDialog("Couldn't open the codec for\"" + *filename
>>> + "\".");
>>> avformat_close_input(&formatCtx);
>>> return false;
>>> }
>>>
>>> sampleFormat = codecCtx->sample_fmt;
>>> if(!(sampleFormat == AV_SAMPLE_FMT_U8 || sampleFormat ==
>>> AV_SAMPLE_FMT_U8P ||
>>> sampleFormat == AV_SAMPLE_FMT_S16 || sampleFormat == AV_SAMPLE_FMT_S16P
>>> ||
>>> sampleFormat == AV_SAMPLE_FMT_S32 || sampleFormat == AV_SAMPLE_FMT_S32P
>>> ||
>>> sampleFormat == AV_SAMPLE_FMT_FLT || sampleFormat == AV_SAMPLE_FMT_FLTP
>>> ||
>>> sampleFormat == AV_SAMPLE_FMT_DBL || sampleFormat == AV_SAMPLE_FMT_DBLP))
>>> {
>>> wxGetApp().popUpErrorDialog("\"" + *filename + "\" uses an unsupported
>>> format.");
>>> avcodec_close(codecCtx);
>>> avformat_close_input(&formatCtx);
>>> return false;
>>> }
>>>
>>> planar = false;
>>> if(sampleFormat == AV_SAMPLE_FMT_U8P || sampleFormat ==
>>> AV_SAMPLE_FMT_S16P ||
>>> sampleFormat == AV_SAMPLE_FMT_S32P || sampleFormat ==
>>> AV_SAMPLE_FMT_FLTP || sampleFormat == AV_SAMPLE_FMT_DBLP)
>>> {
>>> planar = true;
>>> }
>>>
>>> AVFrame* frame = av_frame_alloc();
>>> if(!frame)
>>> {
>>> wxGetApp().popUpErrorDialog("Failed to allocate an audio frame.");
>>> avcodec_close(codecCtx);
>>> avformat_close_input(&formatCtx);
>>> return false;
>>> }
>>>
>>> AVPacket packet;
>>> av_init_packet(&packet);
>>>
>>> duration = formatCtx->duration / (double)AV_TIME_BASE; // duration is
>>> defined in AudioDecoder.h as "double duration = 0;"
>>>
>>> uint64_t estimatedBuffSize = std::ceil(duration * codecCtx->sample_rate
>>> * av_get_bytes_per_sample(codecCtx->sample_fmt) * codecCtx->channels);
>>>
>>> sampleBuffer = (uint8_t*)std::malloc(estimatedBuffSize); // sampleBuffer
>>> is defined in AudioDecoder.h as "uint8_t* sampleBuffer = nullptr;"
>>> sampleBufferSet = true;
>>>
>>> int len;
>>> int gotFrame = 0;
>>> uint64_t dataSize;
>>> uint64_t totalBufferSize = 0;
>>> uint64_t totalSamples = 0;
>>> while(av_read_frame(formatCtx, &packet) == 0)
>>> {
>>> len = avcodec_decode_audio4(codecCtx, frame, &gotFrame, &packet);
>>> if(len < 0)
>>> {
>>> wxGetApp().popUpErrorDialog("Error while decoding.");
>>> std::free(sampleBuffer);
>>> sampleBufferSet = false;
>>> av_packet_unref(&packet);
>>> av_frame_free(&frame);
>>> avcodec_close(codecCtx);
>>> avformat_close_input(&formatCtx);
>>> return false;
>>> }
>>>
>>> if(gotFrame)
>>> {
>>> dataSize = av_samples_get_buffer_size(nullptr, codecCtx->channels,
>>> frame->nb_samples, codecCtx->sample_fmt, 1);
>>>
>>> while(totalBufferSize + dataSize > estimatedBuffSize)
>>> {
>>> estimatedBuffSize *= 1.1;
>>> sampleBuffer = (uint8_t*)std::realloc(sampleBuffer, estimatedBuffSize);
>>> }
>>>
>>> std::memcpy(sampleBuffer + totalBufferSize, frame->data[0], dataSize);
>>>
>>> totalBufferSize += dataSize;
>>> totalSamples += frame->nb_samples;
>>> }
>>>
>>> av_packet_unref(&packet);
>>> }
>>>
>>> sampleBuffer = (uint8_t*)std::realloc(sampleBuffer, totalBufferSize);
>>>
>>> numChannels = codecCtx->channels;
>>> sampleRate = codecCtx->sample_rate;
>>> numSamples = totalSamples;
>>> bufferSize = totalBufferSize;
>>>
>>> av_packet_unref(&packet);
>>> av_frame_free(&frame);
>>> avcodec_close(codecCtx);
>>> avformat_close_input(&formatCtx);
>>>
>>> didInit = true;
>>>
>>> return true;
>>> }
>>>
>>> On Fri, Jun 17, 2016 at 4:20 PM, Bill Messenger <apothemmusic at gmail.com>
>>> wrote:
>>>
>>>> Update: I found out that it only crashes in debug mode. When I build it
>>>> in release mode, it doesn't crash. It must be a bug in MSVC 2015 or
>>>> something.
>>>>
>>>> On Fri, Jun 17, 2016 at 4:06 PM, Bill Messenger <apothemmusic at gmail.com
>>>> > wrote:
>>>>
>>>>> I'm trying to create a class that uses FFmpeg to decode any audio file
>>>>> and store it into memory. Then it has a function that returns a float value
>>>>> of any sample in that buffer. The code I wrote works perfectly for wav and
>>>>> flac files, produces weird audio for mp3 and ogg files, and crashes on
>>>>> certain mp3 files. I spent days trying to figure out why it isn't working,
>>>>> but I can't come up with anything.
>>>>>
>>>>> I think the reason why the audio is weird for mp3 and ogg files is
>>>>> that it uses planar audio instead of interleaved audio, but I don't see
>>>>> what's wrong with the code I wrote. I may be missing something though. For
>>>>> example, to get a sample for 16 bit interleaved audio I use:
>>>>>
>>>>> int16_t tmp = ((int16_t*)sampleBuffer)[numChannels*sample + channel];
>>>>> rv = (float)tmp / 32767.0f;
>>>>>
>>>>> and to get a sample for 16 bit planar audio I use:
>>>>>
>>>>> int16_t tmp = ((int16_t*)sampleBuffer)[sample + numSamples*channel];
>>>>> rv = (float)tmp / 32767.0f;
>>>>>
>>>>> And I have no clue why it crashes on certain mp3 files. I paid close
>>>>> attention to make sure there is enough memory allocated in the buffer.
>>>>> What's even weirder is that the file I created "Chiptune 15 2.mp3" didn't
>>>>> crash, but when I renamed it to "test.mp3", it crashed! These crashes
>>>>> happen on line 139 of "AudioDecoder.cpp":
>>>>>
>>>>> std::memcpy(sampleBuffer + totalBufferSize, frame->extended_data[0],
>>>>> dataSize);
>>>>>
>>>>> with an "Access violation reading location" error in
>>>>> vcruntime140d.dll. It says it isn't with location 0x0000000000000000 or
>>>>> 0xFFFFFFFFFFFFFFFF though, it's a different random location.
>>>>>
>>>>> I attached a zip file with the c++ code and two mp3's. Oh yeah, I
>>>>> should also mention that I'm using MSVC 2015 Community in Windows 10.
>>>>>
>>>>
>>>>
>>>
>>
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://ffmpeg.org/pipermail/libav-user/attachments/20160625/79fdc652/attachment.html>
More information about the Libav-user
mailing list