view docs/man3/SDL_OpenAudio.3 @ 3069:caefe2344f65

Date: Thu, 27 Dec 2007 07:38:25 +0000 From: John Bartholomew Subject: [SDL] SDL Semaphore implementation broken on Windows? Hi, Over the past couple of days, I've been battling with SDL, SDL_Mixer and SMPEG to try to find an audio hang bug. I believe I've found the problem, which I think is a race condition inside SDL's semaphore implementation (at least the Windows implementation). The semaphore code uses Windows' built in semaphore functions, but it also maintains a separate count value. This count value is updated with bare increment and decrement operations in SemPost and SemWaitTimeout - no locking primitives to protect them. In tracking down the apparent audio bug, I found that at some point a semaphore's count value was being decremented to -1, which is clearly not a valid value for it to take. I'm still not certain exactly what sequence of operations is occuring for this to happen, but I believe that overall it's a race condition between a thread calling SemPost (which increments the count) and the thread on the other end calling SemWait (which decrements it). I will try to make a test case to verify this, but I'm not sure if I'll be able to (threading errors being difficult to reproduce even in the best circumstances). However, assuming this is the cause of my problems, there is a very simple fix: Windows provides InterlockedIncrement() and InterlockedDecrement() functions to perform increments and decrements which are guaranteed to be atomic. So the fix is in thread/win32/SDL_syssem.c: replace occurrences of --sem->count with InterlockedDecrement(&sem->count); and replace occurrences of ++sem->count with InterlockedIncrement(&sem->count); This is using SDL v1.2.12, built with VC++ 2008 Express, running on a Core 2 duo processor.
author Sam Lantinga <slouken@libsdl.org>
date Tue, 17 Feb 2009 05:39:18 +0000
parents 546f7c1eb755
children 1238da4a7112
line wrap: on
line source

.TH "SDL_OpenAudio" "3" "Tue 11 Sep 2001, 22:58" "SDL" "SDL API Reference" 
.SH "NAME"
SDL_OpenAudio \- Opens the audio device with the desired parameters\&.
.SH "SYNOPSIS"
.PP
\fB#include "SDL\&.h"
.sp
\fBint \fBSDL_OpenAudio\fP\fR(\fBSDL_AudioSpec *desired, SDL_AudioSpec *obtained\fR);
.SH "DESCRIPTION"
.PP
This function opens the audio device with the \fBdesired\fR parameters, and returns 0 if successful, placing the actual hardware parameters in the structure pointed to by \fBobtained\fR\&. If \fBobtained\fR is NULL, the audio data passed to the callback function will be guaranteed to be in the requested format, and will be automatically converted to the hardware audio format if necessary\&. This function returns -1 if it failed to open the audio device, or couldn\&'t set up the audio thread\&.
.PP
To open the audio device a \fBdesired\fR \fI\fBSDL_AudioSpec\fR\fR must be created\&. 
.PP
.nf
\f(CWSDL_AudioSpec *desired;
\&.
\&.
desired=(SDL_AudioSpec *)malloc(sizeof(SDL_AudioSpec));\fR
.fi
.PP
 You must then fill this structure with your desired audio specifications\&.
.IP "\fBdesired\fR->\fBfreq\fR" 10The desired audio frequency in samples-per-second\&.
.IP "\fBdesired\fR->\fBformat\fR" 10The desired audio format (see \fI\fBSDL_AudioSpec\fR\fR)
.IP "\fBdesired\fR->\fBsamples\fR" 10The desired size of the audio buffer in samples\&. This number should be a power of two, and may be adjusted by the audio driver to a value more suitable for the hardware\&. Good values seem to range between 512 and 8192 inclusive, depending on the application and CPU speed\&. Smaller values yield faster response time, but can lead to underflow if the application is doing heavy processing and cannot fill the audio buffer in time\&. A stereo sample consists of both right and left channels in LR ordering\&. Note that the number of samples is directly related to time by the following formula: ms = (samples*1000)/freq
.IP "\fBdesired\fR->\fBcallback\fR" 10This should be set to a function that will be called when the audio device is ready for more data\&. It is passed a pointer to the audio buffer, and the length in bytes of the audio buffer\&. This function usually runs in a separate thread, and so you should protect data structures that it accesses by calling \fI\fBSDL_LockAudio\fP\fR and \fI\fBSDL_UnlockAudio\fP\fR in your code\&. The callback prototype is: 
.PP
.nf
\f(CWvoid callback(void *userdata, Uint8 *stream, int len);\fR
.fi
.PP
 \fBuserdata\fR is the pointer stored in \fBuserdata\fR field of the \fBSDL_AudioSpec\fR\&. \fBstream\fR is a pointer to the audio buffer you want to fill with information and \fBlen\fR is the length of the audio buffer in bytes\&.
.IP "\fBdesired\fR->\fBuserdata\fR" 10This pointer is passed as the first parameter to the \fBcallback\fP function\&.
.PP
\fBSDL_OpenAudio\fP reads these fields from the \fBdesired\fR \fBSDL_AudioSpec\fR structure pass to the function and attempts to find an audio configuration matching your \fBdesired\fR\&. As mentioned above, if the \fBobtained\fR parameter is \fBNULL\fP then SDL with convert from your \fBdesired\fR audio settings to the hardware settings as it plays\&.
.PP
If \fBobtained\fR is \fBNULL\fP then the \fBdesired\fR \fBSDL_AudioSpec\fR is your working specification, otherwise the \fBobtained\fR \fBSDL_AudioSpec\fR becomes the working specification and the \fBdesirec\fR specification can be deleted\&. The data in the working specification is used when building \fBSDL_AudioCVT\fR\&'s for converting loaded data to the hardware format\&.
.PP
\fBSDL_OpenAudio\fP calculates the \fBsize\fR and \fBsilence\fR fields for both the \fBdesired\fR and \fBobtained\fR specifications\&. The \fBsize\fR field stores the total size of the audio buffer in bytes, while the \fBsilence\fR stores the value used to represent silence in the audio buffer
.PP
The audio device starts out playing \fBsilence\fR when it\&'s opened, and should be enabled for playing by calling \fI\fBSDL_PauseAudio\fP(\fB0\fR)\fR when you are ready for your audio \fBcallback\fR function to be called\&. Since the audio driver may modify the requested \fBsize\fR of the audio buffer, you should allocate any local mixing buffers after you open the audio device\&.
.SH "EXAMPLES"
.PP
.nf
\f(CW/* Prototype of our callback function */
void my_audio_callback(void *userdata, Uint8 *stream, int len);

/* Open the audio device */
SDL_AudioSpec *desired, *obtained;
SDL_AudioSpec *hardware_spec;

/* Allocate a desired SDL_AudioSpec */
desired=(SDL_AudioSpec *)malloc(sizeof(SDL_AudioSpec));

/* Allocate space for the obtained SDL_AudioSpec */
obtained=(SDL_AudioSpec *)malloc(sizeof(SDL_AudioSpec));

/* 22050Hz - FM Radio quality */
desired->freq=22050;

/* 16-bit signed audio */
desired->format=AUDIO_S16LSB;

/* Mono */
desired->channels=0;

/* Large audio buffer reduces risk of dropouts but increases response time */
desired->samples=8192;

/* Our callback function */
desired->callback=my_audio_callback;

desired->userdata=NULL;

/* Open the audio device */
if ( SDL_OpenAudio(desired, obtained) < 0 ){
  fprintf(stderr, "Couldn\&'t open audio: %s
", SDL_GetError());
  exit(-1);
}
/* desired spec is no longer needed */
free(desired);
hardware_spec=obtained;
\&.
\&.
/* Prepare callback for playing */
\&.
\&.
\&.
/* Start playing */
SDL_PauseAudio(0);\fR
.fi
.PP
.SH "SEE ALSO"
.PP
\fI\fBSDL_AudioSpec\fP\fR, \fI\fBSDL_LockAudio\fP\fR, \fI\fBSDL_UnlockAudio\fP\fR, \fI\fBSDL_PauseAudio\fP\fR
...\" created by instant / docbook-to-man, Tue 11 Sep 2001, 22:58