Module Name: src
Committed By: jmcneill
Date: Sun Aug 15 22:32:02 UTC 2010
Modified Files:
src/sys/dev/pci/hdaudio: hdaudio_afg.c
Log Message:
kern/43542: 32kHz playback sounds like hissing analog TV snow
Instead of specifying a range of supported sample rates based on the lowest
and highest available frequencies, only advertise rates that the codec reports
as working. Let aurateconv take care of the rest.
To generate a diff of this commit:
cvs rdiff -u -r1.24 -r1.25 src/sys/dev/pci/hdaudio/hdaudio_afg.c
Please note that diffs are not public domain; they are subject to the
copyright notices on the relevant files.
Modified files:
Index: src/sys/dev/pci/hdaudio/hdaudio_afg.c
diff -u src/sys/dev/pci/hdaudio/hdaudio_afg.c:1.24 src/sys/dev/pci/hdaudio/hdaudio_afg.c:1.25
--- src/sys/dev/pci/hdaudio/hdaudio_afg.c:1.24 Sun Aug 15 19:39:56 2010
+++ src/sys/dev/pci/hdaudio/hdaudio_afg.c Sun Aug 15 22:32:02 2010
@@ -1,4 +1,4 @@
-/* $NetBSD: hdaudio_afg.c,v 1.24 2010/08/15 19:39:56 jmcneill Exp $ */
+/* $NetBSD: hdaudio_afg.c,v 1.25 2010/08/15 22:32:02 jmcneill Exp $ */
/*
* Copyright (c) 2009 Precedence Technologies Ltd <supp...@precedence.co.uk>
@@ -60,7 +60,7 @@
*/
#include <sys/cdefs.h>
-__KERNEL_RCSID(0, "$NetBSD: hdaudio_afg.c,v 1.24 2010/08/15 19:39:56 jmcneill Exp $");
+__KERNEL_RCSID(0, "$NetBSD: hdaudio_afg.c,v 1.25 2010/08/15 22:32:02 jmcneill Exp $");
#include <sys/types.h>
#include <sys/param.h>
@@ -110,6 +110,11 @@
#define HDAUDIO_HP_SENSE_PERIOD hz
+const u_int hdaudio_afg_possible_rates[] = {
+ 8000, 11025, 16000, 22050, 32000, 44100,
+ 48000, 88200, 96000, 176500, 192000, /* 384000, */
+};
+
static const char *hdaudio_afg_mixer_names[] = HDAUDIO_DEVICE_NAMES;
static const char *hdaudio_afg_port_connectivity[] = {
@@ -2903,9 +2908,10 @@
static bool
hdaudio_afg_probe_encoding(struct hdaudio_afg_softc *sc,
- u_int minrate, u_int maxrate, u_int validbits, u_int precision, bool force)
+ u_int validbits, u_int precision, bool force)
{
struct audio_format f;
+ int i;
if (!force && hdaudio_afg_bits_supported(sc, validbits) == false)
return false;
@@ -2919,8 +2925,11 @@
f.channels = 0;
f.channel_mask = 0;
f.frequency_type = 0;
- f.frequency[0] = minrate;
- f.frequency[1] = maxrate;
+ for (i = 0; i < __arraycount(hdaudio_afg_possible_rates); i++) {
+ u_int rate = hdaudio_afg_possible_rates[i];
+ if (hdaudio_afg_rate_supported(sc, rate))
+ f.frequency[f.frequency_type++] = rate;
+ }
#define HDAUDIO_INITFMT(ch, chmask) \
do { \
@@ -2953,17 +2962,11 @@
static void
hdaudio_afg_configure_encodings(struct hdaudio_afg_softc *sc)
{
- const u_int possible_rates[] = {
- 8000, 11025, 16000, 22050, 32000, 44100,
- 48000, 88200, 96000, 176500, 192000, /* 384000, */
- };
struct hdaudio_assoc *as = sc->sc_assocs;
struct audio_format f;
- u_int minrate, maxrate;
int nchan, i;
sc->sc_pchan = sc->sc_rchan = 0;
- minrate = maxrate = 0;
for (nchan = 0, i = 0; i < sc->sc_nassocs; i++) {
nchan = hdaudio_afg_assoc_count_channels(sc, &as[i],
@@ -2979,34 +2982,24 @@
}
hda_print(sc, "%dch/%dch", sc->sc_pchan, sc->sc_rchan);
- for (i = 0; __arraycount(possible_rates); i++)
- if (hdaudio_afg_rate_supported(sc, possible_rates[i])) {
- minrate = possible_rates[i];
- break;
- }
- for (i = __arraycount(possible_rates) - 1; i >= 0; i--)
- if (hdaudio_afg_rate_supported(sc, possible_rates[i])) {
- maxrate = possible_rates[i];
- break;
- }
- KASSERT(minrate > 0 && maxrate > 0); /* impossible */
- hda_print1(sc, " %uHz", minrate);
- if (minrate != maxrate)
- hda_print1(sc, "-%uHz", maxrate);
+ for (i = 0; i < __arraycount(hdaudio_afg_possible_rates); i++)
+ if (hdaudio_afg_rate_supported(sc,
+ hdaudio_afg_possible_rates[i]))
+ hda_print1(sc, " %uHz", hdaudio_afg_possible_rates[i]);
- if (hdaudio_afg_probe_encoding(sc, minrate, maxrate, 8, 16, false))
+ if (hdaudio_afg_probe_encoding(sc, 8, 16, false))
hda_print1(sc, " 8/16");
- if (hdaudio_afg_probe_encoding(sc, minrate, maxrate, 16, 16, false))
+ if (hdaudio_afg_probe_encoding(sc, 16, 16, false))
hda_print1(sc, " 16/16");
- if (hdaudio_afg_probe_encoding(sc, minrate, maxrate, 20, 32, false))
+ if (hdaudio_afg_probe_encoding(sc, 20, 32, false))
hda_print1(sc, " 20/32");
- if (hdaudio_afg_probe_encoding(sc, minrate, maxrate, 24, 32, false))
+ if (hdaudio_afg_probe_encoding(sc, 24, 32, false))
hda_print1(sc, " 24/32");
- if (hdaudio_afg_probe_encoding(sc, minrate, maxrate, 32, 32, false))
+ if (hdaudio_afg_probe_encoding(sc, 32, 32, false))
hda_print1(sc, " 32/32");
if (sc->sc_audiodev.ad_nformats == 0) {
- hdaudio_afg_probe_encoding(sc, minrate, maxrate, 16, 16, true);
+ hdaudio_afg_probe_encoding(sc, 16, 16, true);
hda_print1(sc, " 16/16*");
}
