Nick Hibma wrote:
>
> The following patch removes the stutter when you Ctrl-Z mpg123. The
> soundcard in my desktop machine at work does not have the stutter and
> that one is a ES1371 card. The patch makes the ess_intr routine look
> more like the sb_intr and es_intr ones.
I just tried this patch, and I've got the ESS 1868. It seems to help
that nasty skipping problem with RealPlayer 5.0, and it definitely
stops the skipping in mpg123. I would consider committing this.
To test out if RealPlayer is skipping or not, try playing any of the
clips at cdnow.com. (They've got samples.) All the net congestion
made rvplayer go crazy before. Now, it's much more well-behaved.
> Index: sb.c
> ===================================================================
> RCS file: /home/ncvs/src/sys/dev/sound/isa/sb.c,v
> retrieving revision 1.49
> diff -u -r1.49 sb.c
> --- sb.c 2000/01/10 03:22:28 1.49
> +++ sb.c 2000/02/10 18:27:14
> @@ -360,7 +360,7 @@
> sb_wr(sb, SBDSP_RST, 0);
> if (sb_get_byte(sb) != 0xAA) {
> DEB(printf("sb_reset_dsp 0x%lx failed\n",
> - rman_get_start(d->io_base)));
> + rman_get_start(sb->io_base)));
> return ENXIO; /* Sorry */
> }
> if (sb->bd_flags & BD_F_ESS)
> @@ -583,14 +583,10 @@
> * We are transferring data in DSP normal mode,
> * so clear the dl to indicate the DMA is stopped.
> */
> - if (sb->pch.buffer->dl > 0) {
> - sb->pch.buffer->dl = -1;
> + if (sb->pch.buffer->dl > 0)
> chn_intr(sb->pch.channel);
> - }
> - if (sb->rch.buffer->dl > 0) {
> - sb->rch.buffer->dl = -1;
> + if (sb->rch.buffer->dl > 0)
> chn_intr(sb->rch.channel);
> - }
> }
>
> static int
- Donn
To Unsubscribe: send mail to [EMAIL PROTECTED]
with "unsubscribe freebsd-current" in the body of the message
