Messages in this thread |  | | Date | Sat, 27 Jul 1996 22:17:18 -0400 (EDT) | From | "Kevin D. McCormick" <> | Subject | Re: Sundry repeating 2.0.x problems |
| |
On 27 Jul 1996, Steven L Baur wrote:
> Since 2.0.0 (This wasn't happening in 1.99.x or earlier) I have been > getting messages like this one. > > Jul 18 08:58:58 deanna kernel: Out of memory for xemacs. > > It generally seems to occur when XEmacs is starting up, and seems > spurious because there is always >30MB of available core+swap when > it dies. I've been over XEmacs enough to be sure this is not an > XEmacs problem.
I've been seeing this sort of thing too. I'm running 2.0.0 now, and one evening when I was doing a whole lot of sound work I started seeing
/dev/dsp: Out of memory
when I had a couple dozen MB free when I would start up my sound program. This happened while *opening* the sound device.
Also, as of kernel 2.0.8, the sound driver misbehaves sometimes when playing at high sample rates on my SB16. The following bit of code has worked fine for me ever since the 1.2 kernels, but in 2.0.8 the sound plays at half speed.
int dsp; int SampRate=44100;
dsp=open("/dev/dsp",O_WRONLY); if (dsp==-1) { printf("Can't open audio device.\n"); exit(1); } if (ioctl(dsp,SNDCTL_DSP_SPEED,&SampRate)==-1) { printf("Can't set audio speed.\n"); exit(1); } i=1; if (ioctl(dsp,SNDCTL_DSP_STEREO,&i)==-1) { printf("Can't set stereo mode.\n"); exit(1); } i=16; if (ioctl(dsp,SNDCTL_DSP_SAMPLESIZE,&i)==-1) { printf("Can't set 16 bit mode.\n"); exit(1); }
(My machine is a SuperMicro P55CMS 100MHz Pentium, 32 MB ram, synchronous cache, with IDE and SCSI and a SoundBlaster 16 ASP.)
------------------------------------------------------------------------------- Kevin "Frostbyte" McCormick kmccorm@mit.edu - root@sub-zero.mit.edu Massachusetts Institute of Technology You should be running Linux! Q: How many Microsoft engineers does it take to screw in a light bulb? A: None. They just define darkness as an industry standard.
|  |